Protective Jurisdiction

Protective Jurisdiction, Aggregate Litigation, and the Limits of Article III
James E. Pfander
Introduction During the past sixty years, theories of protective jurisdiction have attracted a good deal of scholarly attention.1 These theories posit that Congress may confer federal question jurisdiction on the federal courts to hear state law claims, even though the claims themselves neither incorporate an original federal ingredient nor seek to enforce rights conferred by federal law.2 But despite ongoing scholarly interest,3 the
Copyright James E. Pfander. Professor, Northwestern University School of Law. Thanks to Paul Mishkin for his many contributions to the field of federal jurisdiction and to Boalt Hall for organizing a symposium in his honor. Thanks as well to Steve Bundy, Steve Gensler, Marty Redish, and George Rutherglen for comments on an earlier draft. 1. Classic early accounts by Herbert Wechsler, Federal Jurisdiction and the Revision of the Judicial Code, 13 Law & Contemp. Prob. 216 (1948) and Paul Mishkin, The Federal Question in the District Courts, 53 Colum. L. Rev. 157 (1953) [hereinafter Mishkin, The Federal Question], have been explored and extended in subsequent work. See Carole E. Goldberg, The Protective Jurisdiction of the Federal Courts, 30 UCLA L. Rev. 542 (1983). For a summary and critique, see Richard H. Fallon, Jr., Daniel J. Meltzer & David L. Shapiro, Hart & Wechslers The Federal Courts and the Federal System 846-55 (5th ed. 2003); Martin H. Redish, Federal Jurisdiction: Tensions in the Allocation of Federal Power 90-95 (2d ed. 1990). 2. Article III extends the judicial power of the United States to cases arising under the Constitution, laws, and treaties of the United States. See U.S. Const., art. III, 2. 28 U.S.C. 1331 gives the federal district courts original jurisdiction of all such civil actions. Scholars conventionally refer to these proceedings as federal question cases and conventionally note that the grant of power in Article III extends more broadly than the general grant of federal question jurisdiction in 1331. See, e.g., Mishkin, The Federal Question, supra note 1, at 160-63. Issues of protective jurisdiction typically arise from the interpretation of other jurisdictional statutes; 1331 has been interpreted to require the existence of a substantial federal question on the face of the well-pleaded complaint. Id. at 164. According to one prominent account, Congress may regulate as it usually does by adopting rules of substance or it may simply transfer the responsibility for applying state law from the state to the federal courts. See Wechsler, supra note 1, at 224-25; see also Eric J. Segall, Article III as a Grant of Power: Protective Jurisdiction, Federalism, and the Federal Courts, 54 Fla. L. Rev. 361, 365-66, 37778 (2002) (approving of the exercise of jurisdiction where Congress simply confers jurisdiction on the federal courts to hear state law claims in an area that it could substantively regulate under its Article I powers). Professor Wechslers support for a broad conception of protective jurisdiction comports with his view that the political safeguards of federalism adequately protect the interests of the states from
1423
1424
CALIFORNIA LAW REVIEW
[Vol. 95:1423
Supreme Court has been notably reluctant to go along. When Justice Jackson first broached the subject in his plurality opinion in the Tidewater case,4 six Justices were quick to disavow his approach.5 In 1957, in perhaps the leading judicial consideration of the subject, Justice Frankfurter rejected two leading theories of protective jurisdiction.6 Since then, the Court has had more than one occasion to embrace protective jurisdiction and has declined to do so.7 Although the Courts apparent rejection of protective jurisdiction suggests a commitment to the classic view of Article III as imposing limits on federal judicial power, the Court has been notably unpredictable in enforcing similar limits in at least two other contexts. To begin, the Court has essentially avoided the question of what limits Article III places on the scope of diversity jurisdiction. As a matter of statutory interpretation, the Court has preserved the traditional rule of complete diversity for many claims brought in federal court.8 But in State Farm Fire & Cas. Co. v. Tashire, the Court upheld Congresss power to relax the complete diversity rule in the interpleader context, treating the issue as too obvious to warrant
congressional encroachments. See Herbert Wechsler, The Political Safeguards of Federalism: The Role of the States in the Composition and Selection of the National Government, 54 Colum. L. Rev. 543 (1954); cf. Jesse J. Choper, Judicial Review and the National Political Process: A Functional Reconsideration of the Role of the Supreme Court (1980). 3. Protective jurisdiction provided an early predicate for some complex litigation proposals. For an account, see Linda S Mullenix, Complex Litigation Reform and Article III Jurisdiction, 59 Fordham L. Rev. 169, 178-91 (1990) (describing ABA task force on mass torts and ALI Project on Complex Litigation as relying upon theories of protective jurisdiction to support jurisdictional expansion). Some have tried to solve the jurisdictional puzzles implicit in the Alien Tort Statute from a protective jurisdiction perspective. See William Casto, The Federal Courts Protective Jurisdiction Over Torts Committed in Violation of the Law of Nations, 18 Conn. L. Rev. 467 (1986). 4. See National Mut. Ins. Co. v. Tidewater Transfer Co., 337 U.S. 582, 586-88 (1949). 5. See Tidewater, 337 U.S. at 617-18 (Rutledge, J., concurring); id. at 647, 652 (Frankfurter, J., dissenting); id. at 628-29 (Vinson, CJ., dissenting). For a critique of, and alternative to, Justice Jacksons suggested approach to the Tidewater problem, see James E. Pfander, The Tidewater Problem: Article III and Constitutional Change, 79 Notre Dame L. Rev. 1925-1980 (2004). 6. See Textile Workers Union v. Lincoln Mills, 353 U.S. 448, 460 (1957) (Frankfurter, J., dissenting). 7. In Verlinden B.V. v. Central Bank of Nigeria, 461 U.S. 480 (1983), the Court found that the claim in question arose under the federal law of foreign sovereign immunity and saw no reason to reach the issue of protective jurisdiction. Still later, in Mesa v. California, the Court pointedly declined the federal governments invitation to adopt protective jurisdiction as the basis for district courts to exercise removal jurisdiction over state law claims brought against federal officials. See Brief for Petitioners at 43, Mesa v. California, 489 U.S. 121 (1989) (No. 87-1206) (describing the federal officer removal statute as meant to protect federal officers from hostile state courts by providing for a trial on the merits of state-law questions free from local interests or prejudice). Instead, the Court chose to limit officer removal jurisdiction to cases in which the federal officers tender substantial defenses grounded in federal law. 8. See, e.g., Exxon Mobil Co. v. Allapattah Servs. Inc., 545 U.S. 546 (2005) (relaxing the amount-in-controversy rule but preserving the complete diversity rule); Owen Equipment & Erect. Co. v. Kroger, 437 U.S. 365 (1978) (expanding the scope of ancillary jurisdiction but preserving the complete diversity rule).
2007]
LIMITS OF ARTICLE III
1425
careful analysis or explication.9 Since Tashire, Congress has come to rely on minimal diversity, rather than protective jurisdiction, as the basis for expanding federal jurisdiction to address certain issues of federal concern.10 The Class Action Fairness Act of 2005 (CAFA), for example, provides for broad federal jurisdiction over multi-state class actions involving claims based upon state law and does so on the basis of diversity between any member of the plaintiff class and the defendants.11 While the Court has yet to rule on the issue, most observers assume that Tashire portends an approval of these new jurisdictional provisions.12 Similarly, the Court has done little to define limits on the scope of a district courts supplemental jurisdiction under Article III. The leading case, United Mine Workers v. Gibbs, holds that a district court may hear a pendent state law claim between non-diverse parties so long as the claim has been joined with a federal question claim over which the court has jurisdiction and arises from the same common nucleus of operative fact[s] as the federal claim.13 But Gibbs does not explore the constitutional boundaries of supplemental jurisdiction in cases where the plaintiff adds pendent parties to the litigation or where diversity of citizenship provides the jurisdictional foundation or anchor. While the Court has approved the exercise of supplemental jurisdiction in the diversity context, it has been careful to preserve the complete diversity rule, at least on the surface.14 The Courts most recent supplemental jurisdiction decision keeps up appearances; Exxon Mobil Corp. v. Allapattah Services, Inc. insists that the complete diversity rule continues to apply to parties in a state law class action.15 Yet in truth, the
9. See State Farm Fire & Cas. Co. v. Tashire, 386 U.S. 523 (1967). 10. Two statutes in particular illustrate the trend. The Multi-party, Multiforum Trial Jurisdiction Act of 2002 (MMTJA) seeks to sweep into federal court all claims arising from any single disaster at a discrete location (such as a fire or plane crash) that results in the death of more than seventy-five individuals. Pub. L. No. 107-273, 116 Stat. 1758 (codified as amended in 28 U.S.C.). More significantly, the Class Action Fairness Act of 2005 (CAFA), Pub. L. No. 109-2, 119 Stat. 4 (codified in 28 U.S.C.), empowers the district courts to hear many state law class actions in which the aggregate value of the claims exceeds $5 million and diversity of citizenship exists between any member of the class and any defendant. See 28 U.S.C. 1332(d)(2). Parts II and IV analyze the MMTJA and CAFA in some detail. 11. Pub. L. No. 109-2, 119 Stat. 4 (codified in 28 U.S.C.). 12. See, e.g., Stephen C. Yeazell, Overhearing Part of a Conversation: Shutts as a Moment in a Long Dialogue, 74 U.M.K.C. L. Rev. 779, 784 n.29 (2006). See generally Charles Alan Wright & Mary Kay Kane, The Law of Federal Courts 159 (6th ed. 2002) (applauding the advent of minimal diversity in Tashire); Erwin Chemerinsky, Federal Jurisdiction 295 (4th ed. 2003) (supporting the use of minimal diversity to overcome the pernicious effects of the complete diversity rule). For a more skeptical view of minimal diversity, see C. Douglas Floyd, The Limits of Minimal Diversity, 65 Hastings L.J. 613 (2004). 13. United Mine Workers v. Gibbs, 383 U.S. 715, 725 (1966). 14. See, e.g., Owen Equipment & Erect. Co. v. Kroger, 437 U.S. 365 (1978). The complete diversity rule holds that none of the properly joined plaintiffs may share a state of citizenship with any properly joined defendant. 15. See Exxon Mobil Corp. v. Allapattah Servs. Inc., 545 U.S. 546 (2005).
1426
[Vol. 95:1423
rule that governs the determination of the citizenship for a plaintiff class, with its focus on the citizenship of the named plaintiff only, actually represents a fairly dramatic departure from a regime of complete diversity. This Article explores the limits of Article III. Part I reviews the rules that govern federal question jurisdiction, the arguments in favor of protective jurisdiction, and the Courts reluctance to permit the gradual erosion of jurisdictional boundaries. Part II examines theories of minimal diversity, showing that such theories could easily circumvent any restrictions the Court has placed on protective jurisdiction. Part III looks at the doctrine of supplemental jurisdiction, now codified in 28 U.S.C. 1367, and explores its tendency to expand jurisdictional boundaries. Part IV draws out the lessons of the earlier discussion. It shows that the shift from state to federal court of state law class actions, though seemingly consistent with current notions of minimal diversity, would represent a profound departure from the Courts jurisprudence of limits. Part IV explores the ways in which the Court might attempt to maintain some coherence in the law that governs the breadth of the judicial power under Article III. In particular, the Court might work to narrow the gap between what Congress can achieve through protective jurisdiction and jurisdiction predicated on minimal diversity of citizenship. I Protective Jurisdiction and the Federal Question To see a role for a theory of protective jurisdiction, one must understand the scope of Article IIIs grant of jurisdiction over cases arising under the Constitution, laws, and treaties of the United States.16 In the leading case, Osborn v. Bank of United States, Chief Justice Marshall held that claims arise under federal law for jurisdictional purposes so long as a federal question forms an ingredient of the original cause.17 An act of Congress incorporated the Bank of the United States and gave it (along with other incidents of corporate personality) a right to sue and be sued in the federal courts. Marshall viewed the Banks capacity to sue as a federal corporation (albeit one in which private parties owned shares) as forming an original ingredient in any suit the Bank might bring or defend, even one brought to enforce the Banks private contractual rights based upon general principles of law. Marshall found that even such private litigation would arise under the federal law of the Banks creation and satisfy Article III.18
16. U.S. Const., art. III, 2, cl. 1. 17. Osborn v. Bank of the United States, 22 U.S. (9 Wheat.) 738, 823 (1824). 18. See generally Anthony J. Bellia, Jr., Article III and the Cause of Action, 89 Iowa L. Rev. 777 (2004) (exploring the Osborn ingredient rule from the perspective of pleading rules and concluding that then-existing rules of pleading would have required corporate entities to plead their capacity to sue). Today, in a world that treats the general principles of contract law to which Marshall referred as matters of state law, the Osborn decision enables Congress to extend federal question jurisdiction to state law
2007]
1427
The problems with the expansive use of Osborns original ingredient test come into sharpest focus in cases in which the parties do not dispute the federal issue and the case turns entirely on state law.19 To be sure, in Osborn itself, state litigants were likely to contest the legality of the Bank and its capacity to sue, making these original ingredients plausible candidates for actual litigation. But by the time of the railroad cases, Congress power to create federal corporations to operate in interstate commerce was well settled.20 Allegations about the capacity of a federal corporation to sue may have appeared as elements in a standard claim for relief, but were unlikely to trigger any litigation. Instead the cases were likely to turn entirely on general law.21 Recognizing that the original ingredient test could result in the transfer of state law matters to federal
claims brought by and against federal instrumentalities or entities. Perhaps the most expansive application of the Osborn rule came in the Pacific Railroad Removal Cases, 115 U.S. 1 (1885), which authorized federal question removal of common law claims against privately owned but federally chartered railroads. For doubts about the continuing viability of those decisions, see Textile Workers Union v. Lincoln Mills, 353 U.S. 448, 470-72 (1957) (Frankfurter, J., dissenting). Congress has limited the application of the instrumentality rule to claims by and against corporations in which the federal government owns a 50% stake. See 28 U.S.C. 1349. Osborn-style ingredient jurisdiction thus helps to explain both the power of the FDIC to sue in federal court as the representative of a failed bank on state causes of action, see DOench Duhme v. FDIC, 315 U.S. 447, 455 (1942) (treating suits by and against the FDIC as suits arising under federal law to which federal common law would apply) and the power of the federal bankruptcy trustee to bring suit in federal court on state law claims against the estates debtors. See Ralph Brubaker, On the Nature of Federal Bankruptcy Jurisdiction: A General Statutory and Constitutional Theory, 41 Wm & Mary L. Rev. 743 (2000) (treating the federal bankruptcy estate as a federal entity under Osborn with power to bring and defend the estates state law claims in federal court). 19. At the time of the Osborn decision, the Court viewed issues of contact and tort law in the bank litigation as matters of general law that it was competent to shape and interpret in light of federal interests. See Osborn, 22 U.S. at 838-46 (rejecting Ohios contention that equity would not restrain a trespass by drawing on general principles and without treating state court decisions as controlling authority); see also Swift v. Tyson, 41 U.S. (16 Pet.) 1 (1842) (applying general common law principles to a dispute over commercial law and rejecting the proposed reliance on state rules of decision). A finding of federal jurisdiction thus entailed the power to fashion controlling rules of common law. Today, jurisdiction no longer implies lawmaking authority. The decision in Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78 (1938), proclaims the end of general federal common law and requires federal courts to treat state judicial decisions as controlling on the substantive issues to which they apply. Lacking power to fashion general common law as an incident of jurisdiction, federal courts may fashion specific common law but only where they identify an authoritative source of positive federal law or an overriding federal interest that empowers them to do so. See Paul J. Mishkin, The Variousness of Federal Law: Competence and Discretion in the Choice of National and State Rules of Decision, 105 U. Pa. L. Rev. 797 (1957). See generally Jay Tidmarsh & Brian J. Murray, A Theory of Federal Common Law, 100 Nw. L. Rev. 585 (2006). Eries conception of the constitutional underpinnings of the state court role in fashioning state law lies at the heart of the Courts suspicion of jurisdictional grants that simply transfer state law matters to federal court; federal courts can resolve disputes over state law, but cannot shape its content as they do with federal law. 20. See Pacific Railroad Removal Cases, 115 U.S. 1 (1885) (upholding removal jurisdiction over common law claim on the theory that the defendant railroad was a federally-chartered corporation under Osborn). 21. Justice Frankfurter described situations in which the federal issue had receded into the remote background of litigation and thus raised doubts about its substantiality. See Textile Workers Union v. Lincoln Mills, 353 U.S. 448, 470-71 (1957).
1428
[Vol. 95:1423
court on the basis of relatively remote federal questions, scholars began to explore more general theories of federal jurisdiction that better explained the scope of congressional power.22 A. Three Theories of Protective Jurisdiction
Three theories of protective jurisdiction have emerged from these efforts to probe the scope of congressional power. First, Professor Wechsler took the position in an early article that Congress should be seen as having the power to provide for federal jurisdiction over any state law claim for which Congress could, in the exercise of its enumerated powers, legislate the rule of decision.23 As a consequence, where Congress by the commerce power could regulate contracts of a certain kind, it would under Wechslers theory be free to take the lesser step of drawing suits upon such contracts to the district courts without displacing substantive state law.24 In this way, Congress might regulate simply by transferring the responsibility for adjudication from state to federal court, without enacting substantive law. Or as Professors Bickel and Wellington put it, providing a forum for the enforcement of state law in a field which Congress could occupy is itself a species of regulation, a way of seeking a degree of uniformity while leaving the maximum room for the exercise of initiative by the states.25 On these views, state law claims would arise under a federal statute that simply conferred jurisdiction on the federal courts. Professor Mishkins theory of protective jurisdiction was less capacious in some ways and more so in others.26 Rather than permitting Congress to grant federal court jurisdiction over any area that Congress could regulate, Mishkin would have required that Congress had actually engaged in some regulation. He was critical of the idea that Congress could simply confer jurisdiction on federal courts to hear claims involving particular parties who might otherwise face state court bias; he viewed the heads of party-alignment jurisdiction in Article III (including diversity
22. Professor Mishkin persuasively criticizes any jurisdictional test that would make original, as opposed to appellate, federal question jurisdiction turn on an estimation as to whether the federal legal issues were substantially contested or disputed. While the parties can identify contested issues of federal law on appeal, original jurisdiction depends on the state of things as of the time of the complaints initial filing. See Mishkin, The Federal Question, supra note 1, at 169-170. Cf. Grable & Sons Metal Products Inc. v. Darue Engineering & Mfg., 545 U.S. 308 (2005) (describing the test for original federal question jurisdiction as depending on the existence of a substantial and disputed issue of federal law). 23. Wechsler, supra note 1, at 224-25. 24. Id. at 225. 25. See Alexander Bickel & Harry Wellington, Legislative Purpose and the Judicial Process: The Lincoln Mills Case, 71 Harv. L. Rev. 1, 20 (1957). 26. Thus, in commenting on the Tidewater decision, Professor Mishkin observed that those who viewed Article III as limiting Congresss power to expand the jurisdiction of the federal courts had the better of the argument on grounds of doctrine, history and authority. See Mishkin, The Federal Question, supra note 1, at 191-92.
2007]
1429
matters) as exhaustively specifying Congresss authority to address such bias.27 It was state hostility against federal programs that Mishkins theory would address. Thus, where Congress has an articulated and active federal policy regulating a field, the arising under clause . . . apparently permits the conferring of jurisdiction on the national courts of all cases in the areaincluding those substantively governed by state law.28 To Mishkin, Osborn and the bankruptcy cases effectively uphold an assertion of protective jurisdiction: Congress provided for federal jurisdiction at least in part to safeguard federal programs from the vagaries of state court adjudication. In distinguishing his approach from Wechslers, Mishkin noted that even under todays expansive conception of the commerce power, Congress might lack the power to govern the substance of the Banks various legal obligations.29 Professor Goldberg extended protective jurisdiction scholarship along three dimensions. First, she offered a definition: protective jurisdiction would apply to claims within federal jurisdiction that rest entirely on state law (and do not qualify for diversity jurisdiction).30 It would not come into play in cases where jurisdiction rests on the presence of federal ingredients in the plaintiffs cause of action, as Osborn implies. Instead of viewing Osborn as an example of protective jurisdiction, Goldberg saw it (and the bankruptcy cases) as an instance of conventional federal ingredient jurisdiction.31 Second, Goldberg criticized what Frankfurter had called Wechslers greater power theory32 and Mishkins partial occupation theory.33 Echoing Frankfurter, Goldberg noted that Congress does not always have the greater power to regulate the particulars of a field of commerce; moreover, there may be too much state-to-state variation and too little agreement in Congress to set forth a detailed federal code.34 In these situations, incorporation of state law may provide the only practical means by which Congress can regulate. Goldberg observed that protective jurisdiction might reflect a congressional desire to secure procedural harmonization or simplification through a shift of litigation to the federal
27. See Mishkin, The Federal Question, supra note 1, at 192. Cf. Goldberg, supra note 1, at 587-89 (questioning Mishkins use of the limits of diversity jurisdiction as an argument for curtailing reliance on protective jurisdiction). Party alignment jurisdiction refers to diversity of citizenship and other provisions of Article III that confer jurisdiction on the basis of the configuration of the parties. 28. Id. 29. Id. at 189. 30. See Goldberg, supra note 1, at 546-50. 31. Id. at 547-48. 32. In rejecting Wechslers analysis in Lincoln Mills, Justice Frankfurter stated that the restrictions of Article III were not met or respected by a beguiling phrase that the greater includes the lesser. Textile Workers Union v. Lincoln Mills, 353 U.S. 448, 474 (1957) (Frankfurter, J. dissenting). 33. Goldberg, supra note 1, at 576-83. 34. Id. at 576-83, 593.
1430
[Vol. 95:1423
system.35 Finally, Goldberg noted concerns with state autonomy, just emerging when she wrote but better established today.36 Her theory of protective jurisdiction would recognize broad federal power to confer protective jurisdiction, subject to the possibility that state autonomy concerns may outweigh federal interests.37 B. Illustration of Competing Theories: Section 301 of the Labor Management Relations Act The much-debated power of Congress to provide for the enforcement of labor contracts illustrates these competing theories of protective jurisdiction. Section 301 of the Labor Management Relations Act (LMRA) confers jurisdiction on the federal courts to hear suits for violation of a contract between an employer and a labor organization.38 While the provision apparently grew out of Congresss desire to secure the judicial enforcement of collective bargaining agreements, particularly in suits against unions, doubts arose about the source of governing law and the legitimacy of the resulting jurisdictional grant. On one commonly held view of the statute, Congress failed to create any substantive body of labor contract law but simply conferred jurisdiction on the federal courts in the expectation that state rules of contract law would apply. All three theories of protective jurisdiction would apparently uphold the jurisdictional grant, so conceived. Most observers would agree that Congress had power to specify the rules that govern collective bargaining agreements between unions and employers in industries affecting commerce. Even without specifying substantive rules, on Wechslers view Congress might take the lesser step of shifting litigation to federal courts, perhaps to address perceived state court bias. Mishkins theory would also provide support for the jurisdictional statute. Although Congress has not specified rules for enforcement of collective agreements in particular, it has actively regulated the field of industrial relations and collective bargaining. Federal law controls the
35. In recognizing the difficulty that regulation can pose for a Congress divided on matters of substance and in identifying the federal interest in procedural coordination as one that might justify a grant of federal jurisdiction, Goldberg anticipated issues close to the heart of Congresss decision to use minimal diversity as the tool of federal jurisdictional expansion in CAFA. See id. at 577-78 (describing the Consumer Class Action Act of 1969 as having relied on state law in part to avoid the necessity of legislating into place a detailed consumer protection code). 36. Id. at 595-601 (citing National League of Cities v. Usery, 426 U.S. 833 (1976)). On the refinement and expansion of the state autonomy rule in later cases, see infra notes 67-68. 37. Professor Goldbergs embrace of a balancing test to determine when Congressional expansion of protective jurisdiction invades state autonomy does not fit especially well with the Rehnquist Courts emphasis in formalistic rules of federal-state relations. See infra notes 67-68 and accompanying text. 38. See 29 U.S.C. 185 (2000). For an account of the passage of the labor contract provisions of the LMRA, see James E. Pfander, Judicial Purpose and the Scholarly Process: The Lincoln Mills Case, 69 Wash. U.L.Q 243 (1991).
2007]
1431
process by which employees choose union representation and imposes duties on both the employer and the union to bargain in good faith to reach an agreement. Within this actively regulated field, Mishkins approach would allow Congress to confer jurisdiction over certain disputes without actually specifying controlling rules for every case. For Goldberg, the statute presented something of a puzzle. Its provision conferring capacity on unions to sue and be sued could be regarded as an original federal ingredient in every union lawsuit, thereby making Osborn-style jurisdiction available.39 But Goldberg would treat the existence of overlap between state and federal legal rules as depriving the federal law of independent force as an original ingredient capable of supporting the exercise of Osborn jurisdiction.40 Because state laws often afford unions the capacity to sue and be sued, Goldberg concluded that the federal union capacity rule cannot establish a distinctive primary right, and cannot function as a source of original ingredient jurisdiction.41 On this view, section 301 should be analyzed as if it conferred protective jurisdiction on the federal courts. Goldberg would uphold the jurisdictional grant so long as the federal interest outweighed the threat to state control over the meaning of state law.42 While she criticized congressional motivation in adopting section 301 as inappropriate,43 she recognized elsewhere that the ability of state courts to exercise concurrent jurisdiction may lessen the threat to their control over state law. The Court side-stepped all of these protective jurisdictional questions in Textile Workers Union v. Lincoln Mills, ruling that section 301 created a federal substantive right to the enforcement of collective bargaining agreements and obliged the federal courts to define the contours of that substantive right as a matter of federal common law.44 On such a view, suits for violation of the collective agreement arose under federal law and
39. See Goldberg, supra note 1, at 562. 40. Id. at 565 (discussing federal incorporation of state consumer protection law). 41. Id. at 562. I agree with this aspect of Goldbergs analysis, but disagree with its application to the labor contract arena. As a general matter, federal incorporation of state law should not suffice to make that law federal for purposes of establishing the federal substance needed to support a grant of federal question jurisdiction. Such incorporation has a bootstrapping quality, transforming otherwise applicable state law into federal law solely for the purpose of facilitating federal jurisdiction but not altering in any respect the nature of the underlying substantive obligations. But if the statute contemplates a change in the substantive law, then incorporation of some state content should present no problem. In the labor contract arena, Congress actually created a uniform federal rule entitling unions to sue and be sued as juristic entities without regard to the rules that would obtain under state law. So while this federal entity status may have paralleled the juristic status that unions would enjoy in some states, the federal entity rule would not depend on incorporated state law and would not vary from state to state. 42. Id. at 604-05, 609-14 43. Id. at 613. 44. See Textile Workers Union v. Lincoln Mills, 353 U.S. U.S. 448, 457 (1957).
1432
[Vol. 95:1423
presented no protective jurisdictional issues.45 Justice Frankfurter, however, thought the statute was better read as a simple grant of jurisdiction that called upon the federal courts to apply state law.46 He thus faced the constitutional issue and the arguments for protective jurisdiction that had been advanced in support of the statute. Writing only for himself, Frankfurter concluded that the statute exceeded the bounds of Article III.47 His opinion represents the most searching judicial consideration of the doctrine of protective jurisdiction and his insistence on enforcing the constitutional limits of Article III appears to have nicely anticipated the Courts current approach. Frankfurter acknowledged that federal ingredients were the basis for jurisdiction in Osborn and the bankruptcy and railroad removal cases.48 But he expressed doubt as to the growing power of the ingredient theory and in any case found no substantial federal ingredient in a section 301 claim.49 Frankfurter then reviewed theories of protective jurisdiction and found them wanting as well. As for Wechslers greater power theory, Frankfurter echoed the view that claims cannot arise under a jurisdictional statute, even one adopted within an area of commerce over which Congress could exercise substantive control.50 Such a theory could vastly extend the jurisdiction of the federal courts to include every contract and tort affecting interstate commerce.51 Equally troubling, such an expansive view would rest on a belief in the inadequacy of state courts in determining state
45. Justices Burton and Harlan concurred, expressing the view that the statute did not authorize the federal courts to fashion federal common law but could nonetheless be upheld as a grant of protective jurisdiction. See Lincoln Mills, 353 U.S. at 459-60 (Burton, J., concurring) (citing Intl Bro. of Teamsters v. W.L. Mead, Inc., 230 F.2d 576 (1956)). 46. Elsewhere, I have questioned Justice Frankfurters conclusion as a matter of statutory interpretation. See Pfander, supra note 38, at 298-304 (collecting evidence that supports the majoritys conclusion that Congress viewed claims for breach of the labor contract as arising under federal, not state, substantive law). 47. Lincoln Mills, 353 U.S. at 484. 48. See id. at 470-72 (citing Osborn v. Bank of the United States, 22 U.S. (9 Wheat.) 738 (1824) and Pacific Railroad Removal Cases, 115 U.S. 1 (1885)). 49. Frankfurter regarded those decisions as having pressed Article III to its very limits, and perhaps beyond. Id. at 470-72. He therefore sought to confine the scope of the jurisdiction to matters involving some substantial federal interest. Bankruptcy matters implicated such an interest as did the desire of Congress to protect the Bank as an instrument of federal policy. Id. at 481-83. But Frankfurter viewed the interests at stake in Lincoln Mills as less significant. Congress had acted to ensure that unions would be treated as a juristic entity for purposes of enforcing their legal obligations, including their collective bargaining agreements. Id. at 480-81. While one might treat this congressional grant of legal personality as an original ingredient in every action the union might bring, comparable to the situation in Osborn, Frankfurter saw an important difference. The Bank was more entirely the creature of its act of incorporation than were the unions covered by section 301. In any case, times had changed: Other modes of invoking federal jurisdiction, including removal by defendants, made the exercise of original jurisdiction on the basis of remote federal ingredients more difficult to defend. Id. at 481-82. 50. Lincoln Mills, 353 U.S. at 474 (Frankfurter, J. dissenting). 51. Id.
2007]
1433
law, a premise inconsistent with Frankfurters conception of the proper role of federal courts in the federal system. As for Mishkins partial occupation theory, Frankfurter was similarly unconvinced. Although the Mishkin approach would apply only to claims within a field that Congress had previously regulated, it would still transfer state law claims to federal court and expand federal power.52 In a telling comment that builds on Mishkins insight, Frankfurter noted that the party-alignment grants of jurisdiction confer a species of protective jurisdiction and argued that only in those situations may Congress shift state law matters into federal court for protective reasons.53 Since Lincoln Mills, the Court has studiously avoided any reliance on the doctrine of protective jurisdiction. One opportunity to consider the doctrine came in connection with the Foreign Sovereign Immunities Act (FSIA), a federal statute that regulates the manner in which individuals may pursue claims against foreign sovereigns.54 The FSIA codifies the so-called restrictive theory of foreign sovereign immunity under which foreign nations may be subject to liability in the courts of the United States in connection with commercial activities that have some impact in this country.55 For citizens of the United States, the FSIA poses no jurisdictional problem; Article III extends federal jurisdiction on the basis of party alignment to controversies between a foreign nation and citizens of one of the (United) States. In Verlinden B.V. v. Central Bank of Nigeria, party-alignment jurisdiction was unavailable; the plaintiff was a Dutch corporation and the defendant was a foreign nation.56 Jurisdiction thus depended on a finding that the claims at issue (for anticipatory repudiation of a letter of credit) arose under federal law. In deciding that question, the Court described its cases (with perhaps more hope than accuracy) as firmly establish[ing] that Congress may not expand the jurisdiction of the federal courts beyond the bounds established by the Constitution.57 As part of this concern with limits, the Court expressly declined to consider the argument that the claims could be sustained as an aspect of so-called protective jurisdiction.58 Rather than rely on protective jurisdiction, the Court emphasized the substantive content of federal law. On the Courts view, the FSIA established a comprehensive federal regulatory scheme that determined not
52. Id. at 474-76. 53. Id. at 476 (quoting Mishkin, The Federal Question, supra note 1, at 192). 54. 90 Stat. 2891-98 (codified at 28 U.S.C. 1330 and elsewhere). 55. For accounts of the operation of the FSIA and a review of its jurisdictional puzzles, see Mary Kay Kane, Suing Foreign Sovereigns: A Procedural Compass, 34 Stan. L. Rev. 385 (1982); Note, The Theory of Protective Jurisdiction, 57 N.Y.U.L. Rev. 933 (1982). 56. See Verlinden B.V. v. Central Bank of Nigeria, 461 U.S. 480 (1983). 57. Id. at 491. 58. Id. at n.17.
1434
[Vol. 95:1423
only the existence of subject matter and personal jurisdiction over claims against foreign defendants but also the availability of an exception to foreign sovereign immunity that would clear the way for the imposition of substantive liability.59 In stressing the Acts regulation of substantive liability, the Court declined to view the FSIA as merely jurisdictional. Every claim under the Act would confront a threshold question as to the amenability of the foreign sovereign to suit and federal law would unquestionably control that question. Once an exception to immunity has been held to apply, the Act declares that the foreign state shall be liable in the same manner as a private individual under like circumstances.60 While the Court viewed this provision as incorporating state law, the foreign sovereign immunity rule was seen as an ingredient of any such claim. The Court thus applied the Osborn rule, upholding federal jurisdiction on the theory that a title or right set up by the party, may be defeated by one construction of [federal law] and sustained by the opposite construction.61 The Courts most recent brush with the doctrine came in Mesa v. California and it produced, if anything, a more forceful insistence on the need for limits and a more skeptical view of protective jurisdiction.62 Two U.S. postal employees cited for traffic violations under California state law removed the state proceedings to federal court under the federal officer removal statute, 28 U.S.C. 1442. The federal government argued that the statute was best read to allow removal anytime the state proceeding targeted action taken by federal employees under color of office. Such jurisdiction might be sustained through a theory of protective jurisdiction even though the officers had failed to tender any federal defense to liability. The Court disagreed. Section 1442 could not independently support federal jurisdiction.63 Citing the need to avoid the serious constitutional questions, the Court held that the statute authorizes removal by federal officers only when they tender a colorable federal defense to liability.64 The Court explained that it had not previously adopted a theory of protective jurisdiction and saw no reason to do so in the Mesa case. Any
59. Id. at 496-97. 60. 28 U.S.C. 2674. 61. Verlinden B.V., 486 U.S. at 492. Critics of the Verlinden decision have portrayed it as purely jurisdictional, questioning the Courts conclusion that substantive liability turned on federal law. See, e.g., Segall, supra note 2, at 380-81. But even as the statutes exception to foreign sovereign immunity keys the exercise of federal subject matter jurisdiction and personal jurisdiction, it also provides a crucial trigger for substantive liability. If a plaintiff cannot find a statutory exception to immunity, the foreign sovereign would enjoy a federal immunity from suit that would control even in a state or federal court that otherwise enjoyed subject matter and personal jurisdiction over the cause. 62. Mesa v. California, 489 U.S. 121 (1989). 63. Id. at 136. 64. Mesa, 489 U.S. at 121-22.
2007]
1435
federal interests were adequately protected by allowing removal in cases involving federal defenses. Like Verlinden, Mesa was unanimous.65 Since Mesa came down in 1988, the Court has had no occasion to address protective jurisdiction.66 But the trajectory of its federalism rulings since Mesa give little reason to suppose the Court has developed a more spacious view of the scope of arising-under jurisdiction. In the past fifteen years, the Court has fashioned new limits on the power of Congress, marginally cutting back on its power to regulate matters affecting interstate commerce and its power to implement its regulatory initiatives through the agency of the states.67 It has also taken a narrow view of Congresss power to permit individuals to sue states to enforce federal regulatory schemes.68 These cases often focus on the importance of identifying a limit to congressional power, a view closely aligned with Justice Frankfurters
65. Justices Brennan and Marshall joined the majority opinion but concurred to address the separate issue that would arise if the government employees had based removal on a well-founded claim of state court bias or hostility. 66. The Courts most recent analysis of the Westfall Act concludes that state-law claims against federal employees arise under federal law for Article III purposes, so long as the Attorney General of the United States certifies that the claims arose from actions taken within the scope of employment. See Osborn v. Haley, 127 S. Ct. 881 (2007). For the Court, certification as to the scope of employment (which in turn bears on federal law issues of immunity and indemnity) provides a federal ingredient sufficient to ground the assertion of federal jurisdiction, even in the unusual case in which the federal court were ultimately to conclude that the certification was unwarranted and state law continued to control liability. As the Court saw matters, certification sets the stage for potential changes in the substantive law: it provides for the substitution of the United States as a defendant, it transforms applicable law by making the Federal Tort Claims Act applicable, and it introduces the possibility of a substantive federal defense of official immunity from suit. Cf. Gutierrez de Martinez v. Lamagno, 515 U.S. 417 (1995) (dissenting opinion) (raising doubts about the constitutionality of Westfall Act under Article III on the basis that its provision for judicial review of a threshold scope-of-employment issue was merely jurisdictional and could not provide the substantial federal ingredient needed to open federal courts to claims otherwise governed by state law). Neither case discussed the possibility of protective jurisdiction. 67. See United States v. Lopez, 514 U.S. 549 (1996) (invalidating federal law that criminalized drug possession as having exceeded congressional power to regulate commerce); United States v. Morrison, 529 U.S. 598 (2000) (invalidating federal statute creating a private right of action for genderbased violence as exceeding the scope of the congressional power over commerce); cf. Gonzales v. Raich, 545 U.S. 1 (2005) (upholding federal power to criminalize private cultivation and use of marijuana for medicinal purposes). See also New York v. United States, 505 U.S. 144 (1992) (invalidating federal statute that commandeered state legislatures into adopting state law that incorporated federal standards); Printz v. United States, 521 U.S. 898 (1997) (extending anti-commandeering principle to federal laws aimed at securing state administrative enforcement of federal statutes). 68. See Alden v. Maine, 527 U.S. 706 (1999) (invalidating federal statute adopted under the commerce clause to the extent that it authorized individuals to sue a state for damages in state court); Seminole Tribe v. Florida, 517 U.S. 44 (1996) (holding that the Eleventh Amendment bars Congress from authorizing individuals to sue states in federal court to enforce a federal commerce statute); cf. Fitzpatrick v. Bitzer, 427 U.S. 445 (1976) (upholding congressional power to abrogate state sovereign immunity through legislation enacted pursuant to the Fourteenth Amendment); Central Virginia Community College v. Katz, 126 S.Ct. 990 (2006) (concluding that the constitutional grant of plenary power enables Congress to subject states to suit in bankruptcy proceedings).
1436
[Vol. 95:1423
criticisms of protective jurisdiction and with the skepticism expressed in Verlinden and Mesa. We can briefly summarize the lessons of the Courts reluctance to embrace protective jurisdiction in the following terms. Congress cannot simply transfer state law claims to the federal courts because it has protected an area of federal concern by empowering the federal courts to apply state law. Cases do not arise under jurisdictional statutes for purposes of satisfying Article III.69 To authorize the federal courts to exercise federal question jurisdiction, Congress must establish federal substantive rights on which the success of the claims in some sense must depend. Congress may declare the content of federal substantive law itself, of course, or, as in the majoritys account of labor contract enforcement under section 301, direct the federal courts to do so. The Courts willingness to fashion federal common law at Congresss behest may depend in part on the degree of guidance Congress has provided.70 Federal entities like the Bank, created by an act of Congress, may bring suit in federal court, even to enforce state law rights of action, because their very existence depends on federal law.71 But a simple declaration that, within a certain area of federal concern, federal courts should hear state law claims or that federal law incorporates state law without altering its content, neither warrants the creation of federal common law nor establishes any rules of federal substance under which claims may arise.72 Similar limits may apply to the exercise of supplemental jurisdiction, as the next Part explores. II Supplemental Jurisdiction and Article III Like protective jurisdiction, supplemental jurisdiction operates to expand the jurisdiction of the federal courts to include state law claims between non-diverse parties over which the district courts lack any independent source of jurisdiction. As with protective jurisdiction, the Court has generally declined to embrace the broadest possible interpretation of the scope of supplemental jurisdiction. Thus, when a claim arising under federal law anchors or grounds the jurisdiction of the
69. See Mesa v. California, 489 U.S. 121, 136 (1989); Verlinden B.V. v. Central Bank of Nigeria, 461 U.S. 480, 496 (1983). 70. The Courts handling of the alien tort statute reveals the limits of the federal common law solution. See Sosa v. Alvarez-Machain, 542 U.S. 692 (2004) (concluding that the alien tort statute lawfully conferred jurisdiction on the federal courts to entertain suits for torts in violation of the law of nations on the theory that such actions would implicate federal common law, but articulating a restrictive view of the torts that would qualify under such an approach). 71. See Am. Natl Red Cross v. S.G., 505 U.S. 247 (1992) (upholding federal jurisdiction over a state law tort claim brought against the Red Cross, a federally-chartered charitable organization). 72. See Lincoln Mills, 353 U.S. at 472-73 (Frankfurter, J. dissenting). Part IV.C below more closely examines the incorporation of state law.
2007]
1437
district courts, the Court has permitted the exercise of supplemental jurisdiction only over those state law claims that satisfy the common nucleus test of United Mine Workers v. Gibbs.73 When the anchor claim itself arises under state law, and jurisdiction depends on diversity of citizenship, the Court has displayed a consistent concern with preserving the complete diversity rule. Because much of the law has developed as a matter of statutory, rather than constitutional interpretation, it remains uncertain today whether the Court would approve an interpretation of supplemental jurisdiction that would erode the requirement of complete diversity. A. Federal Questions and Supplemental Jurisdiction
Some scholars who have traced the origins of supplemental jurisdiction have noted that the doctrine developed along different paths on the federal question and diversity side of the federal docket. On the federal question side, cases have focused on the power of the district court to hear multiple claims, one based on federal law and another similar claim based on state law (against a non-diverse defendant). Although early decisions required a fairly close relationship between the two claims, the Warren Court significantly relaxed these requirements. In the leading case, United Mine Workers v. Gibbs,74 the Court announced that the scope of a constitutional case within the meaning of Article III extends to a federal question claim over which the district court has original jurisdiction and to a non-federal (state law) claim that arises from the same common nucleus of operative facts.75 Gibbs has been widely viewed as permitting the assertion of jurisdiction over claims, one federal and one state, which meet modern procedural tests of transactional relationship. That, indeed, seemingly explains the Courts reasoning that the two claims form a part of the same constitutional case for jurisdictional purposes. The evolution of supplemental jurisdiction from pendent claims to pendent parties occasioned a bump or two in the road. In Aldinger v. Howard, the Court refused to permit the assertion of pendent party jurisdiction over an additional defendant where no independent basis of federal jurisdiction existed.76 While the federal statute in question had been interpreted to authorize suit against county officials, the statute was not then viewed as authorizing similar federal actions against the county itself. The plaintiff nonetheless sued the county under state law in federal court
73. 383 U.S. 715 (1966). 74. Id. 75. Id. at 725. In Gibbs, the Court permitted a plaintiff to join a federal question claim for a secondary boycott with a business tort claim under state law. Both claims arose from the same labor dispute, and easily satisfied the common nucleus test, even though they did not meet the more demanding same cause of action standard of earlier cases. 76. 427 U.S. 1 (1976).
1438
[Vol. 95:1423
and argued that the state law claim arose from the same common nucleus of operative facts as the federal claim against the county official. While the Gibbs common nucleus test was doubtless satisfied, the Court refused to permit the assertion of supplemental jurisdiction. That refusal, coupled with the restrictive views expressed in the Courts subsequent decision in Finley v. United States,77 raised substantial doubts about the availability of pendent party jurisdiction and led to the adoption of the supplemental jurisdiction statute.78 B. Diversity Jurisdiction and Aggregate Litigation Meanwhile, on the diversity side of the docket, pendent jurisdictional concepts had little traction. Rules of aggregation had long since permitted a single plaintiff to join all claims (even unrelated claims) against a diverse defendant: no transactional relationship among the claims was required and the doctrine of pendent claim jurisdiction had no application.79 Moreover, in cases of multi-party litigation, Strawbridge v. Curtiss barred jurisdiction on the basis of diversity unless all of the plaintiffs were citizens of states different from those of the defendants.80 Even where the pendent partys claims bore a close relationship to those of other diverse litigants, the complete diversity rule foreclosed the assertion of jurisdiction over the new non-diverse party.81 As a consequence, development in the law of supplemental jurisdiction on the diversity side of the federal docket came in connection with the assertion not of pendent jurisdiction but of ancillary jurisdiction. Although the concept defies easy definition, ancillary jurisdiction extends to claims that bear such a relationship to a controversy properly before the court that they cannot, in fairness to an interested party, be excluded from the litigation unit.82 Ancillary jurisdiction began as an outgrowth of the old rule that state courts may not entertain claims to property over which a federal court has acquired control in the course of litigation.83 Thus, when a dispute between diverse parties brings a specific piece of property or fund before the federal court, federal control of the property was seen as displacing the power of state courts to hear related
77. 490 U.S. 545 (1989). 78. James E. Pfander, Supplemental Jurisdiction and Section 1367: The Case for a Sympathetic Textualism, 148 U. Pa. L. Rev. 109 (1999) [hereinafter Pfander, Sympathetic Textualism]. 79. Id. 80. Strawbridge v. Curtiss, 7 U.S. (3 Cranch) 267 (1806) (articulating the complete diversity rule). 81. Pfander, Sympathetic Textualism, supra note 78. 82. See Wright & Kane, supra note 12, at 34. 83. See, e.g., Freeman v. Howe, 65 U.S. (24 How.) 450 (1861). On the evolution of ancillary jurisdiction after Freeman, see Richard A. Matasar, Rediscovering One Constitutional Case: Procedural Rules and the Rejection of the Gibbs Test for Supplemental Jurisdiction, 71 Calif. L. Rev. 1399, 1463-66 (1983).
2007]
1439
and possibly competing claims to the same property. This displacement of state court control, in turn, necessitated an assertion of ancillary jurisdiction over claims by non-diverse parties to the same property or fund. The resulting theory of ancillary jurisdiction over property reached its high-water mark with the development of the equity receivership.84 Later cases extended ancillary jurisdiction beyond the property context.85 Yet the leading modern case, Owen Equipment & Erection Co. v. Kroger, attempts to preserve the distinction between the separate and distinct claims of pendent parties (to which the complete diversity rule still applies) and claims within ancillary jurisdiction (which could be heard in federal court).86 The Kroger Court confirmed that the controversy between the plaintiff (Kroger) and a diverse defendant (OPPD) could expand to support the exercise of ancillary jurisdiction over third-party impleader claims against Owen, and over other claims such as interpleader and intervention as of right.87 But the Court refused to permit the assertion of jurisdiction over the plaintiff Krogers controversy with the non-diverse defendant, Owen. Ancillary jurisdiction was said to require more than factual similarity, it required logical dependence.88 Joinder of separate, if related, claims may serve an interest in economic and efficient adjudication, but those interests did not override the complete diversity rule
84. For background on the equity receivership, see Rosenberg, Swaine & Walker, Corporate Reorganization and the Federal Court (1924); David McC. Wright, Jurisdiction and Venue in Federal Equity Receivership of Corporations, 24 Va. L. Rev. 29 (1937-38). In such a proceeding, a creditor would bring suit in equity against a corporation (often a railroad company) setting up an unpaid debt and the companys inability to pay. Once the federal court was satisfied that the creditors claims were sound, it could appoint a receiver to take over the operation of the assets (often real property) of the company. Once appointed, the receiver would often pursue claims in federal court against the companys debtors, acting in some respects like a trustee in bankruptcy. The Court held that these receiver suits were ancillary to the initial bill in equity that led to the appointment of the receiver. For a sense of the power of the receivership as a vehicle of jurisdictional expansion, see White v. Ewing, 159 U.S. 36 (1895) (jurisdiction acquired over action in equity on the basis of diversity to secure the appointment of receiver; once receiver was appointed and assets of the firm were brought before the court, ancillary jurisdiction extended to claims by against the corporation in receivership without regard to diversity). The Court commented in passing that the jurisdiction in an equity receivership, based on diversity, does not differ materially from that of the district court in bankruptcy. Ewing, 159 U.S. at 40. 85. See, e.g., Moore v. New York Cotton Exchange, 270 U.S. 593 (1926) (ancillary jurisdiction extended to a state common law claim asserted by defendant as a compulsory counterclaim to a federal antitrust action). 86. See Owen Equipment & Erection Co. v. Kroger, 437 U.S. 365 (1978). Kroger brought a diversity action against the Omaha Power District (OPPD) seeking damages for the wrongful death of her husband. The power company initiated a third-party impleader claim for contribution or indemnity, alleging that Owen Equipment bore responsibility for any damages that might be awarded against the power company. Kroger amended her complaint to assert a direct claim against Owen, despite the fact that the two parties were not diverse. 87. See id. at 376 n.18 (approving of prior decisions that exercised ancillary jurisdiction over compulsory counterclaims, interpleader claims, cross claims, and claims seeking intervention of right). 88. Id. at 376.
1440
[Vol. 95:1423
for pendent party controversies, especially when the state courts were available to hear all related claims.89 C. Modern Supplemental Jurisdiction: Exxon Mobile Co. v. Allapattah Servs. Inc. While the Kroger Court was careful to place its decision on statutory grounds,90 its insistence on the preservation of the complete diversity rule continues to inform modern practice. The supplemental jurisdiction statute, which was drafted twelve years after Kroger, took pains in section 1367(b) to preserve Kroger and the rules of complete diversity.91 And the Courts most recent interpretation of the statute works hard to deliver on that promise of preserved complete diversity. In Exxon Mobil v. Allapattah Servs. Inc. the Court finally resolved a long-running debate over the application of the supplemental jurisdiction statute to class actions based upon state law.92 That debate had featured conflicting claims about the operation of supplemental jurisdiction in cases based on diversity of citizenship. On one literal view, section 1367(a) had abrogated the complete diversity rule by permitting the exercise of Gibbs style pendent party jurisdiction in diversity proceedings. For literalists, the failure of section 1367(b) to create an exception for the claims of plaintiffs joined under Rule 20 or 23 of the Federal Rules of Civil Procedure resulted in a broad expansion of federal jurisdiction. On the other hand, some contended that section 1367(a) implicitly incorporated the rules of original jurisdiction that had previously controlled the joinder of additional parties in diversity and thus foreclosed the exercise of supplemental jurisdiction except where the traditional complete diversity rule had been satisfied.93 On this more conservative view, the rules of complete diversity continued to control and the doctrine of ancillary jurisdiction, as approved and limited in Kroger, continued to define the boundaries of supplemental jurisdiction in diversity matters. In Exxon Mobil, the Court found a way to straddle the two positions. As for diversity of citizenship, the Court concluded that the statute preserved existing law. Thus, supplemental jurisdiction will operate in diversity only where the alignment of the parties satisfies the complete diversity rules as they have been defined. (On that issue, the otherwise divided Court was unanimous.) As for the amount-in-controversy requirement, however, the Court joined the literal camp. Once the parties
89. Id. at 376 (noting the availability of the state court as a forum in which the plaintiffs claims could be heard). 90. Id. at 374 n.13 (describing as settled the conclusion that the complete diversity rule is not a constitutional requirement). 91. Pfander, Sympathetic Textualism, supra note 78. 92. 545 U.S. 546 (2005). 93. Pfander, Sympathetic Textualism, supra note 78.
2007]
1441
have been aligned so as to satisfy complete diversity and at least one party has asserted a claim that satisfies the statutorily defined jurisdictional threshold (now $75,000), district courts may assert supplemental jurisdiction over the claims of additional parties whose claims satisfy the common-nucleus test of Gibbs and section 1367(a). Thus, the Court agreed that the district court had jurisdiction over a state law class action where the representatives of the plaintiff class satisfied the complete diversity rule and met the amount-in-controversy threshold. Supplemental jurisdiction was available over the other claimants in the plaintiff class, even though they claimed damages below the threshold amount. In a companion case, the Court applied the same rule to plaintiffs joined under Rule 20; so long as one plaintiff satisfied the statutory amount-in-controversy threshold, supplemental jurisdiction extended to other diverse plaintiffs with related (but less substantial) claims.94 Exxon Mobil stops well short of suggesting that the diversity requirement derives from the Constitution. Indeed, as we shall see in the next Part, convention holds that Congress can relax the complete diversity rule and authorize federal jurisdiction on the basis of minimal diversity of citizenship between opposing parties. But without a constitutional underpinning, the Exxon Mobil Courts insistence upon complete diversity does not make a great deal of sense. The literal reading of the supplemental jurisdiction statute that the Court embraced in relaxing the amount-incontroversy requirement would seemingly require similar relaxation of the complete diversity rule. Yet the Court painstakingly preserved its complete diversity rule, at some cost to its claim of straightforward text-based interpretation. As the dissent noted, it is quite difficult to distinguish diversity and amount-in-controversy requirements as a matter of statutory interpretation; both requirements appear in the text of section 1332 and both were threatened by literalism. But Article III of the Constitution does provide a logical basis for the distinction; it says nothing about the amount in controversy but does limit federal jurisdiction to controversies between citizens of different states.95 The next Part explores the rise of minimal diversity and the erosion of constitutional limits. III The Rise of Minimal Diversity Although the Courts refusal to embrace protective jurisdiction and to expand supplemental jurisdiction displays some concern with the preservation of limits, the Court has indicated that Congress has the power to relax the complete diversity rule. While the venerable complete diversity rule of Stawbridge v. Curtiss continues to control the exercise of
94. 95. Rosario-Ortega v. Starkist Foods, Inc., 545 U.S. 546 (2005). U.S. Const., art. III, 2.
1442
[Vol. 95:1423
jurisdiction over most simple disputes,96 the Court has approved jurisdiction over complex multi-party claims on the basis of minimal diversity between adverse parties. In State Farm Fire & Cas. Co. v. Tashire,97 a case that arose from a collision between a truck and a bus, the Court faced questions about the breadth of diversity jurisdiction and about the proper scope of an interpleader action.98 As for the second issue, the Court concluded that the district court had expanded the proceeding beyond the fair limits of interpleader.99 En route to this finding, the Court ruled that the interpleader statute had been correctly and constitutionally interpreted to confer jurisdiction on the federal courts on the basis of minimal diversity.100 It relied in part on lower court authority, on its own decisions in some prior cases, and on the analysis in a well-known study by the American Law Institute (ALI).101 Delivered during the heyday of the Warren Court, Tashire displays little concern with the need for limits. But
96. See Strawbridge v. Curtiss, 7 U.S. (3 Cranch) 267 (1806) (articulating the complete diversity rule). 97. 386 U.S. 523 (1967). 98. There, State Farm brought an interpleader action in federal court, seeking to compel all interested parties to join in litigation over a truck-bus collision in California that killed or injured nearly forty individuals from five different states and Canada. State Farm insured the truck driver, under a policy that limited its liability to $20,000 and obliged the firm to pay defense costs. State Farm hoped to discharge its obligation by paying the policy limits into the court and disclaiming any further duty to defend. Meanwhile, the bus company sought to piggyback on State Farms interpleader action; it sought to require all potential victims to prosecute their tort claims against the bus company and its driver in the interpleader proceeding. Ultimately, the district court agreed to expand the proceeding as requested and granted injunctive relief that barred the injured parties from pursuing their tort claims in other forums. 99. State Farm, 386 U.S. at 533-37. Interpleader was not to serve as an all-purpose bill of peace in which a single accident would give rise to litigation in a single forum with all parties required to participate. Rather, interpleader came into play only when one party could point to a stake or limited fund and a risk of excessive or duplicative liability. That threat existed in the case of State Farm; if litigation were permitted to go forward in a variety of forums, the insurance company might face indemnification obligations that exceeded its policy limits or, more likely, some victims of its insureds negligence might find themselves incapable of sharing in the proceeds of the policy. But the other party in the interpleader proceeding, the bus company, could not claim any similar threat of inconsistent or excessive liability. Though potentially quite substantial, the many personal injury claims against the bus company were not limited or capped as were the claims against State Farm. In the end, State Farm was entitled to limited reliefan order restraining parties from enforcing claims against it except in the context of the interpleader actionand the bus company was denied interpleader relief altogether. 100. The statute itself authorizes jurisdiction whenever a stakeholder brings an action in the nature of interpleader involving assets or property worth at least $500 that may be claimed by two or more adverse claimants of diverse citizenship. See 28 U.S.C. 1335. The jurisdiction thus depends on the existence of diversity between any two adverse claimants to a fund or property, a requirement that was satisfied in the State Farm case by the potential claims of injured passengers from California, Oregon, and Canada. 101. See State Farm, 386 U.S. at 530-31 (citing federal circuit court authority, its earlier decision in Barney v. Latham, 103 U.S. 205 (1880)). The Court also relied on a draft analysis of the constitutionality of minimal diversity that later appeared as an appendix to the final report. See American Law Institute, Study of the Division of Jurisdiction Between State and Federal Courts, Reporters Memorandum A 431 (1969).
2007]
1443
the Courts somewhat casual approach indicates that it did not mean to break new ground. A. The Multi-party, Multiforum Trial Jurisdiction Act
Building on the holding of Tashire, Congress has grown somewhat more assertive about relying upon minimal diversity as a tool for jurisdictional expansion. An initial step came in 2002, with the adoption of the Multi-party, Multiforum Trial Jurisdiction Act (MMTJA).102 The Act provides for federal jurisdiction over claims between citizens of the same state, which grew out of an accident in which seventy-five or more people die. The MMTJA addresses these potentially non-diverse but factually related claims in two ways. First, the Act permits the plaintiff to intervene in any pending district court proceeding involving claims arising from the same accident, as long as the minimal diversity test is satisfied. Second, the Act permits removal of a non-diverse state proceeding to federal court so long as the defendant has become a party to other litigation from the same accident that involves at least one diverse plaintiff. In this way, the MMTJA provides for the assertion of diversity jurisdiction over claims between non-diverse parties so long as the claims arise from a qualifying accident that produces litigation otherwise meeting the minimal diversity test. B. The Class Action Fairness Act of 2005
The desire to achieve coordinated treatment, and other goals, informed the adoption of the Class Action Fairness Act of 2005 (CAFA).103
102. See 28 U.S.C. 1369, 1441(e). The idea of multi-party joinder in complex litigation arose with the publication of the ALI Study in 1969, and gained strength in the ALIs study on complex litigation in 1994. For background, see Thomas D. Rowe, Jr. & Kenneth Sibley, Beyond Diversity: Federal Multi-party, Multiforum Jurisdiction, 135 U. Pa. L. Rev. 7 (1986). The Act seeks to provide for coordinated treatment of litigation that stems from a single accident at a discrete location that claims the lives of at least seventy-five persons. Where the requisite triggering event has occurred, the statute authorizes the district courts to exercise original and removal jurisdiction on the basis of minimal diversity between adverse claimants. Such a jurisdictional grant would surely encompass all airplane accidents and the Kansas City skywalk disaster, given the likely dispersion of citizenship among claimants. The Act assumes that jurisdictional coordination will occur following the invocation of federal jurisdiction through the offices of the Judicial Panel on Multidistrict Litigation (JPML). The JPML may order the transfer of actions for coordinated or consolidated pre-trial proceedings any time civil actions pending in different districts involve one or more common questions of law or fact. The transferee court, sometimes known as a multidistrict litigation or MDL court, then oversees discovery on a coordinated basis. (A drafting glitch resulted in the omission of language from section 1407 that would have authorized the MDL court to retain the cases for trial.) A few years before adopting the MMTJA, Congress adopted the Y2K Act, imposing notice and heightened pleading requirements on class actions brought to secure damages resulting from computer glitches traceable to the Y2K bug. See Pub. L. No. 106-37, 113 Stat. 185 (1999) (codified at 15 U.S.C.) For an analysis of jurisdiction over class actions under the Y2K Act, see part IV below. 103. See Pub. L. No. 109-2, 119 Stat. 4 (codified at 28 U.S.C. 1332(c)-(d), 1453, 1711-15). For background on the statute, and recognition that it proceeds on the basis of minimal diversity, see
1444
[Vol. 95:1423
Although CAFA includes a few substantive provisions that regulate the fairness of class action litigation and settlement, jurisdictional provisions lie at the heart of the Act.104 The Act provides for original jurisdiction over any class action, so long as the aggregate value of the claims of the class members exceeds $5 million and there exists any minimal diversity of citizenship between any member of the plaintiff class and any defendant. The Act thus modifies existing law and expands federal jurisdiction in two respects. First, it determines the citizenship of the class members by reference to their individual citizenship status; the old rule focused on the citizenship of the named plaintiff in at least some situations.105 Second, the Act also permits aggregation of claim value to meet the jurisdictional threshold; the old rule required that the claims of each plaintiff class member satisfy the amount in controversy.106 Although it includes exceptions to preserve state court control over some in-state class actions, the Act shifts a wide range of consumer class actions to federal court.107 The breadth of the jurisdictional provisions reflects Congresss desire essentially to federalize consumer class actions. Among the many justifications for the Act, two loomed large: Congresss desire to address a range of problems associated with state court class action litigation108 and its desire to secure the coordinated treatment of
Edward F. Sherman, Class Actions After the Class Action Fairness Act of 2005, 80 Tulane L. Rev.1593, 1595-96 (2006). For a review and critique of CAFAs apparent assumption that federal courts would decline to apply state choice-of-law rules in deciding whether to certify class actions, see Patrick Woolley, Erie and Choice of Law After the Class Action Fairness Act, 80 Tulane L. Rev. 1723 (2006). 104. CAFA imposes limits on coupon settlements, attorneys fees, and settlements that entail a loss by class members. See 28 U.S.C. 1712-13. But these provisions apply only to class actions that were brought in or removed to federal court. See 28 U.S.C. 1711(2). They do not apparently apply to class actions terminated in state court. See Woolley, supra note 103, at 1751 n.141. 105. See Supreme Tribe of Ben Hur v. Cauble, 255 U.S. 356, 366-67 (1921). Part IV explains that the citizenship rule in Cauble applies to true class actions but does not necessarily apply to Rule 23(b)(3) class actions. 106. See Snyder v. Harris, 394 U.S. 332 (1969); Zahn v. International Paper Co., 414 U.S. 291 (1973); cf. Exxon Mobil Corp. v. Allapattah Servs. Inc., 545 U.S. 546 (2005) (interpreting supplemental jurisdiction statute as abrogating Zahn but not Snyder). 107. The major exceptions are for small-value class actions (worth less than the $5 million value specified in the statute); class actions brought against States or state officials; and class actions involving fewer than 100 members. See 28 U.S.C. 1332(c)(2), 1332(c)(5). In addition, the Act authorizes the district court to decline jurisdiction where many class members and the primary defendants are citizens of single state. See 28 U.S.C. 1332(c)(3)-(4). But a class action made up entirely of plaintiffs from a single state would be subject to federal jurisdiction on the basis of diversity of citizenship between any plaintiff and any primary defendant, even if the remaining primary defendants were from the same state as the plaintiffs. 108. By authorizing state courts to entertain nationwide class actions on the basis of the class members decision, however well considered, to decline to opt out, see Phillips Petroleum Co. v. Shutts, 472 U.S. 797 (1985), the Shutts decision led to a number of perceived problems. For one thing, Shutts opened the doors of every state court to the assertion of nationwide class actions, a form of universal venue that led to forum-shopping by well-informed class action lawyers. Particular state courts gained reputations for their willingness to certify class actions and for awarding damages at trial
2007]
1445
related class actions in federal court.109 By providing for broad original and removal jurisdiction, the Act clears the way for the operation of the Judicial Panel on Multidistrict Litigation (JPML) and the transfer of related litigation to a single multidistrict litigation (MDL) court for pre-trial proceedings that will often include a decision about whether to certify the class.110 CAFA illustrates the breadth of minimal diversity as a device for expanding federal jurisdiction. Indeed, Congress accomplished through CAFA much of what it previously declined to do with a proposed grant of protective jurisdiction. In 1969, Senator Joseph Tydings of Maryland introduced S. 1900, legislation that broadly authorized the federal courts to assert jurisdiction over consumer class actions based on state law.111 The
that threatened the financial health of corporate defendants. See Senate Judiciary Committee Report on Class Action Fairness Act, S. Rep. No. 109-14 at 13-27 (2005) reprinted in 2005 U.S.C.C.A.N. 3, 1630 (describing a series of state class action abuses, including excessive fees for lawyers, inadequate relief for class members, forum shopping that leads to the growth of cases in class action friendly forums like Madison County, Illinois, overlapping and duplicative actions, and nationwide actions in which small, rural counties propose to dictate the regulatory terms that will govern commercial activities in other states, sometimes ignoring the conflicting state laws in those other states). In these bet-your-company situations, defendants could face considerable pressure to settle following a state court certification decision that confronted the defendant with the prospect of significant potential liability to the class as a whole. See Victor E. Schwartz, Federal Courts Should Decide Interstate Class Actions: A Call for Federal Class Action Diversity Reform, 37 Harv. J. on Legis. 483, 499-501 (2000) (describing class certification problems in Alabama); John H. Beisner & Jessica Davidson Miller, Theyre Making a Federal Case Out of It . . . In State Court, 25 Harv. J.L. Poly 143 (2001) (offering a catalog of the problems with state court handling of multi-state class actions). 109. See Robert H. Klonoff, Introduction to the Symposium, 74 U.M.K.C. L. Rev. 487, 490 (2006); Edward F. Sherman, Class Actions After the Class Action Fairness Act of 2005, 80 Tulane L. Rev. 1593, 1595 (2006). 110. MDL courts view the certification issue as one that they may address in class actions transferred for coordinated treatment under 28 U.S.C. 1407. See In re Tyco Intl, Ltd., 236 F.R.D. 62 (D.N.H. 2006) By shifting the class certification decision to federal court, the legislation may make it more difficult to certify nationwide class actions involving claims governed by state law. Unlike some of their state counterparts, federal courts have expressed doubts about the certification of nationwide state law classes under Rule 23(b)(3). For cases on the federal side that take a somewhat dim view of nationwide class actions due in part to conflicting bodies of state law, see In re Am. Med. Sys. Inc., 75 F.3d 1069, 1085 (6th Cir. 1996); In re Rhone-Poulenc Rorer, Inc., 51 F.3d 1293, 1300-01 (7th Cir. 1995). For evidence that Congress understood that a shift of class actions to federal court would trigger this more restrictive approach, see S. Rep. No. 109-14, supra note 108, at 14 (noting that many state court judges are lax about following the strict requirements of Rule 23 but that federal judges pay closer attention to the procedural requirements for certifying a class action). As a result, CAFA may reduce the number of nationwide class actions and lead to more state-wide consumer class actions instead. See Edward H. Cooper, Rewriting Shutts for Fun, Not to Profit, 74 U.M.K.C. L. Rev. 569, 581 (2006). But this result depends on the willingness of federal courts to ignore state choice-of-law rules in certifying nationwide classes. See generally Richard A. Nagareda, Bootstrapping in Choice of Law After the Class Action Fairness Act, 74 U.M.K.C. L. Rev. 661, (2006) (exploring the choice of law rules for class actions and expressing skepticism about the capacity of the class action to achieve law reform goals in the absence of substantive legislation). 111. An early draft of the legislation provided simply that the district court shall have original jurisdiction, regardless of the amount in controversy or the citizenship of the parties, of civil class actions brought by one or more consumers . . . where (1) the action involves the violation of consumers rights under State or Federal statutory or decisional law. S. 1980, 91st Cong., 1st Sess.
1446
[Vol. 95:1423
apparent purpose of the legislation was, as with CAFA, to shift consumer class actions into federal court to secure the application of what were then perceived as the more liberal joinder provisions of Rule 23.112 Supporters of the legislation also noted the advantages of coordinated treatment through the JPML.113 Like CAFA, the 1969 legislation failed to create federal substantive rights for consumers, declaring instead that the federal courts were to exercise jurisdiction over claims based upon state law. But instead of relying upon minimal diversity, the 1969 legislation would have apparently proceeded on a theory of protective jurisdiction or on federal incorporation of state law.114 Despite testimony in favor of the constitutionality of protective jurisdiction from no less a figure than Yale Law School Professor Charles Black, doubts persisted and the legislation eventually died.115 The concerted opposition of the very business groups
2(b), reprinted in 115 Cong. Rec. 10460-61 (1969). Later versions of the legislation expressly declared unlawful an act in defraud of consumers which affects commerce and conferred original jurisdiction on the district courts to entertain civil class actions for redress. See H.R. 14585, 91st Cong., 1st Sess. 4 (1969), reprinted in Class Action and Other Consumer Protection Proceedings, Hearings Before the Subcommittee on Commerce and Finance of the Committee on Interstate and Foreign Commerce, 91st Cong., 2d Sess. at 5 (1970) [hereinafter Consumer Hearings]. This later bill expressly included any act which gives rise to a civil action by a consumer or consumers under State statutory or decisional law. Id. The bill thus proceeded by adopting or incorporating state law as a federal standard and making it clear that the law to be applied would be determined as if the jurisdiction of the Federal court were based on diversity of citizenship. Id. at 6. For an evaluation of the constitutionality of the Tydings legislation, see Note, Federal JurisdictionProtective Jurisdiction and Adoption as Alternative Techniques for Conferring Jurisdiction on Federal Courts in Consumer Class Actions, 69 Mich. L. Rev. 710, 711-13 (1971). 112. See Statement of Sen. Joseph Tydings, Consumer Hearings, supra note 111, at 37 (noting that the bill would secure the application of the liberal machinery of Rule 23); Statement of Rep. Bob Eckhart, id. at 10 (discussing the House companion to the Tydings bill and emphasizing the importance of securing the application of the liberal provisions of Rule 23); cf. Statement of Richard McClaren, US Antitrust Division, reprinted in id. at 200, 206 (commenting that the principal justification for the legislation was the inadequacy of state class action procedures). 113. See Statement of Rep. Eckhart, Consumer Hearings, supra note 111, at 19-20 (describing use of the multi-district litigation machinery for cases involving class actions in more than one district). 114. For an analysis of the Consumer Class Action Act under a protective jurisdictional framework, see Goldberg, supra note 1, at 565, 577-78. For a summary of the Acts two jurisdictional approaches, see note supra. Interestingly, the decision to rely upon a protective jurisdiction/federal question approach resulted from disappointment with the Courts decisions narrowing the availability of diversity jurisdiction. See Statement of Sen Joseph Tydings, Consumer Hearings, supra note 111, at 37 (expressing concern with the Courts restrictive decision in Snyder v. Harris, 394 U.S. 332 (1969)). Today, the push to rely upon minimal diversity stems in part from doubts about the availability of protective jurisdiction. See Richard A. Epstein, The Consolidation of Complex Litigation: A Critical Evaluation of the ALI Proposal, 10 J. L. & Com. 1, 35 (1990-91) (recounting that the ALI complex litigation project shifted from reliance on federal question/protective jurisdiction to minimal diversity jurisdiction as the basis for federal jurisdiction over complex, multi-state litigation). 115. See Statement of Charles Black, Consumer Hearings, supra note 111, at 21 (opining that the early version of the statute would pass constitutional muster as a grant of protective jurisdiction and that the later version would even more clearly do so in light of its incorporation or adoption of state law as a federal rule of decision). For doubts on this score, see Goldberg, supra note 1, at 555-56, 565 n.144; Note, supra note 111, at 729-31.
2007]
1447
that were later to support the adoption of CAFA played a supporting role.116 C. Expanding Federal Jurisdiction through Minimal Diversity
The juxtaposition of CAFA and the proposed 1969 legislation illustrates a larger truth: that Congress can vastly expand federal jurisdiction through minimal diversity and accomplish much of what it might otherwise seek to achieve through protective jurisdiction. Consider first the example of the Osborn case. There, as we have seen, the Court took quite a broad view of the scope of federal question jurisdiction, viewing it as extending to any action that the Bank, as a federal instrumentality, might bring to enforce rights based upon state law. While many, including Justice Frankfurter, have viewed Osborn as pressing the outer boundaries of Article III, Congress might have achieved the same result through minimal diversity. It could simply define a corporation as a citizen of every state in which its shareholders were citizens and permit the corporation to sue and be sued on the basis of any minimal diversity between a shareholder and an opposing party. On such a view, the existence of diversity between Osborn, a citizen of Ohio, and some owner of the Bank would confer jurisdiction.117 If minimal diversity could solve the Osborn problem, at least as to publicly held corporations with dispersed shareholders, it could also overcome most of the nettlesome protective jurisdiction issues that have arisen to date. Take, for example, the Lincoln Mills case.118 As an
116. See Consumer Hearings, supra note 111, at 261, 418-19 (testimony by representatives of the American Retail Federation and the Chamber of Commerce in opposition to the legislation). 117. Some might question the power of Congress to re-define the citizenship of parties for purposes of expanding diversity jurisdiction. After all, corporations have long been treated as citizens of their state of incorporation and principal place of business. See 28 U.S.C. 1332(c)(1) (deeming a corporation to be a citizen of its state of incorporation and its principal place of business). On the origins of this deeming provision in 1958, see Wright & Kane, supra note 12, at 165-68. But at the time of the Osborn decision, the Court had taken a different view. The corporation itself was not a citizen of any state within the meaning of Article III. Instead, the Court looked to the citizenship of the shareholders or owners of the corporation in determining the existence of diversity, much the way the Court today deems the citizenship of unincorporated associations and partnerships to be that of their members. See Bank of the United States v. Deveaux, 9 U.S. (5 Cranch) 61, 67-70 (1809). Only later did the Court establish a conclusive, if fictional, presumption that corporations were citizens of their state of incorporation. See Marshall v. Balt. & Ohio R.R., 57 U.S. (16 How.) 314, 328 (1855); Louisville R.R. Co. v. Letson, 43 U.S. (2 How.) 497, 552-56 (1844). For an account of these developments, see James W. Moore & Donald T. Weckstein, Corporations and Diversity of Citizenship Jurisdiction: A Supreme Court Fiction Revisited, 77 Harv. L. Rev. 1426 (1964); Dudley O. McGovney, A Supreme Court Fiction: Corporations in the Diverse Citizenship Jurisdiction of the Federal Courts, 56 Harv. L. Rev. 853 (1943). Later still, Congress added the corporations principal place of business as an additional state of citizenship. With this history of evolving meaning and congressional involvement, it seems unlikely that a decision by Congress to return to the Courts own original definition of corporate citizenship would raise constitutional eyebrows. 118. See Textile Workers Union v. Lincoln Mills, 353 U.S. 448 (1957).
1448
[Vol. 95:1423
unincorporated association, the Textile Workers Union would have been regarded as a citizen of every state in which its members were citizens.119 Its claim against Lincoln Mills to compel arbitration of a labor dispute under the collective agreement would have surely met the minimal diversity test, however one defines corporate citizenship. In the bankruptcy setting, corporations abound both as plaintiffs and defendants, creditors and debtors. It would be an unusual case indeed that did not involve at least some minimal diversity of citizenship between interested parties. Provisions for removal and intervention modeled upon the multi-party legislation of 2002 would enable the bankruptcy court (or district court) to hear all claims conceivably related to a particular bankruptcy. Indeed, the coordination functions that federal courts perform in the exercise of federal question jurisdiction in bankruptcy bear more than a passing resemblance to those they are now expected to perform in exercising minimal diversity jurisdiction over certain interpleader and class action proceedings. CAFA and MMTJA illustrate the potential breadth of minimal diversity as a device for expanding federal jurisdiction. By using minimal diversity, Congress accomplished through CAFA much of what it previously declined to do with a proposed grant of protective jurisdiction. If the Courts limits on protective jurisdiction are to remain meaningful, the Court must explore ways of reining in the broadest forms of minimal diversity. Part IV begins that exploration. IV The Search for Limits The advent of minimal diversity in such federal laws as the Class Action Fairness Act and the Multi-party, Multiforum Trial Jurisdiction Act suggests a need to rethink the limits of Article III. Restrictions on protective jurisdiction make little sense in a world where Congress can seemingly expand federal jurisdiction on the basis of minimal diversity. Perhaps, then, the Court should reconsider its reluctance to embrace the doctrine of protective jurisdiction. Alternatively, the Court may choose to resist minimal diversity and encourage Congress to articulate federal standards sufficient to support the exercise of federal question jurisdiction. In this Part, the Article explores the range of responses available to a Court that chose to confront inconsistencies in its jurisprudence of limits.
119. See United Steelworkers v. R.H. Bouligny, Inc., 382 U.S. 145 (1965) (treating the citizenship of unions organized as unincorporated associations as defined by the citizenship of their members).
2007]
1449
A.
The Textual and Doctrinal Case for Limits on the Scope of Minimal Diversity
The Court might revisit minimal diversity, narrowing its scope as a tool of jurisdictional expansion in order to maintain its view that Article III imposes limits on the judicial power. The Court might begin such a project with the text of Article III itself, which in relevant part extends the judicial power to controversies between citizens of different states. If we take as a starting point the idea that the term controversy refers to a judicially cognizable dispute,120 it follows that the jurisdictional grant extends only to the disputes that arise between citizens of different states. Use of the term between suggests a head-to-head dispute involving two opposing parties from different states, like one between Hamilton (New York) and Jefferson (Virginia).121 The absence of any reference to the subject matter of the dispute suggests that the cognizability of controversies under Article III depends entirely on the alignment of the parties, not on the nature or subject matter of the dispute.122 That much seems quite straightforward. Matters grow more complicated when the scope of the litigation expands to include additional parties and claims. Suppose that Hamilton (New York) and Madison (Virginia) bring suit against Jefferson (Virginia). One might say that the controversy between Hamilton and Jefferson satisfies the constitutional requirement and anchors the exercise of federal jurisdiction; once anchored, the jurisdiction could conceivably extend to the non-diverse dispute between Madison and Jefferson as well. One might
120. A growing body of scholarship suggests that cases differ from controversies in Article III. See, e.g., Robert J. Pushaw, Jr., Article IIIs Case/Controversy Distinction and the Dual Functions of Federal Courts, 69 Notre Dame L. Rev. 447, 450 (1994) (cases involve the federal courts primarily as expositors of federal law, whereas controversies involve a bilateral dispute wherein a judge served primarily as a neutral umpire whose decision bound only the immediate parties); William A. Fletcher, The Case or Controversy Requirement in State Court Adjudication of Federal Questions, 78 Calif. L. Rev. 263, 266-67 (1990) (collecting evidence that cases include both civil and criminal proceedings, whereas controversies may include only civil matters); James E. Pfander, Rethinking the Supreme Courts Original Jurisdiction in State-Party Cases, 82 Calif. L. Rev. 555 (1994) (same). 121. On the meaning of the word between in the framers lexicon, see William Winslow Crosskey, 1 Politics and the Constitution in the History of the United States 78-82 (1953) (collecting evidence from the founding era that the term between connotes a bilateral relationship and narrows the scope of diversity jurisdiction). See also Pushaw, supra note 120, at 450 (drawing on rich collection of historic materials in defining a controversy as a bilateral dispute in which the judge serves as a neutral umpire). 122. See Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 378 (1821). Consider, for example, the distinctive approach to aggregation of claims that developed on the diversity side of the federal docket. Rules of aggregation permit an individual plaintiff to add together two unrelated claims against a diverse defendant to satisfy the statutory amount-in-controversy requirement; its not the subject matter relationship between such claims that justifies their aggregation but the simple fact that a controversy between diverse citizens meets the statutory threshold. See Pfander, Sympathetic Textualism, supra note 78, at 130 & n.85. By contrast, the joinder of claims on the federal question side of the docket depends on the existence of a transactional relationship between the claims. See text accompanying notes 74-75 (discussing the Gibbs rule authorizing the exercise of pendent jurisdiction over a federal question claim and a state law claim that share the same common nucleus of operative fact).
1450
[Vol. 95:1423
analyze this proposed expansion of the litigation from the perspective of potential bias. By virtue of having joined in litigation with Hamilton, Madison may face some risk of bias in the Virginia courts and justifiably invoke federal jurisdiction. Alternatively, one might say that Madisons presence on the same side with Hamilton eliminates any threat of bias to Hamilton in a Virginia court, especially if the interests of Hamilton and Madison are closely aligned and the jury must rule for them both, up or down.123 Arguments about bias, in short, do not help much in defining constitutional limits on the scope of the litigation unit for diversity purposes.124 An alternative approach to the evaluation of jurisdiction might focus on the nature of Madisons claims that were proposed for addition to the litigation. If those claims form a part (in some appropriate sense) of the Hamilton controversy, then it may make sense to expand federal jurisdiction to reach them. If, by contrast, the claim comprises a separate controversy, then expansion of the litigation unit to reach the claim could present a problem. Article III requires a controversy between citizens of different states: the Hamilton+Madison v. Jefferson litigation may satisfy that test (even absent complete diversity) but the Madison v. Jefferson litigation, standing alone, would not. The challenge lies in determining when one can properly treat the Madison v. Jefferson matter as part of the Hamilton v. Jefferson controversy properly before the court. If the Madison v. Jefferson claim presents a separate controversy, the text of Article III might well be read to require that it stand on its own jurisdictional footing. The problem lies in defining the proper scope of a controversy for purposes of Article III. The Court might decide that it cannot develop an administrable test for the scope of a controversy. Or it might decide that it should simply defer to Congress and the drafters of modern procedural systems.125 But if the Court were inclined to search for limits, one line of decisions long embedded in the law of diversity may suggest a solution. As a matter of statutory interpretation, the Court has generally refused to countenance an expansion of jurisdiction when Madisons claim would be viewed as separate and distinct from that of Hamilton. As the Court explained in an early decision, when two or more plaintiffs, having separate and distinct demands, unite for convenience and economy in a single case, it is essential that the demand of each independently satisfy jurisdictional
123. In Strawbridge v. Curtiss, 7 U.S. (3 Cranch) 267, 267-78 (1806), the Court indicated that all parties who join together in asserting a joint interest must satisfy the diversity requirement. 124. See Henry Friendly, Historic Basis of Diversity Jurisdiction, 41 Harv. L. Rev. 483, 509 (1928) (reaching the same conclusion); ALI Study, supra note 101, at 433 (same conclusion). 125. See Matasar, supra note 83, at 1448-54 (criticizing as too narrow the common nucleus test of Gibbs and suggesting the need for greater deference to modern procedural rules in defining the scope of a case under Article III).
2007]
1451
requirements.126 The requirement of jurisdictional self-sufficiency would mean that Madisons claim must, if separate, satisfy the party-alignment test for diversity of citizenship and exceed any applicable statutory amount-in-controversy threshold. (Under Article III, of course, Congress could reduce or eliminate any amount-in-controversy threshold.)127 If the Court were to characterize Madisons claim as separate and distinct, it might refuse to permit him to rely on Hamiltons claim to furnish the needed diverse citizenship. The text of Article III thus provides a straightforward basis for restricting congressional use of minimal diversity jurisdiction. The remainder of this Part evaluates the cogency of such limits in light of prior decisions that seemingly accept minimal diversity. Careful review of those decisions reveals the surprising truth that prior law does not compel acceptance of the boldest assertions of minimal diversity. 1. Evaluating the Doctrinal Basis for Minimal Diversity Imposing limits on the scope of minimal diversity may appear inconsistent with the Courts prior decisional law and with the existence of ancillary jurisdiction. But commonly cited authorities stop short of endorsing the broad exercise of minimal diversity jurisdiction over separate but related claims of pendent parties. For example, in Barney v. Latham, the Court upheld removal jurisdiction over an action that included separate claims involving both diverse and non-diverse parties.128 But Barney does
126. Troy Bank v. G.A. Whitehead & Co., 222 U.S. 39, 40 (1922) (addressing the separate and distinct issue as it relates to the determination of the amount in controversy). See also Johnson v. Manhattan Railway Co., 289 U.S. 479, 496-97 (1933) (noting that consolidation is permitted as a matter of convenience and economy in administration, but does not merge the suits into a single cause, or change the rights of the parties, or make those who are parties in one suit parties in another). 127. Well developed in the context of the joinder of claims, see Clark v. Paul Gray, Inc., 306 U.S. 583 (1939) (plaintiffs who sue on separate claims must each satisfy the amount-in-controversy requirement), this rule was extended to class actions proposed for certification under Rule 23(b)(3). Thus, the Court held that the members of a (b)(3) class action were asserting separate and distinct claims. See Snyder v. Harris, 394 U.S. 332, 334 (1969). The Courts subsequent decision in Zahn v. International Paper Co., 414 U.S. 1, 15-16 (1973), applied the same rule to class members, even where the named class representative asserted claims in excess of the statutory value. The Court later found that the supplemental jurisdiction statute overruled this aspect of the Zahn decision. See Exxon Mobil Corp. v. Allapattah Servs. Inc., 545 U.S. 546 (2005). 128. See Barney v. Latham, 103 U.S. 205 (1880) (upholding removal jurisdiction under the precursor to 28 U.S.C. 1441(c) which now authorizes a defendant to remove the entire action when the plaintiff joins a federal question claim within the district courts original jurisdiction and a separate and distinct state law claim). The statute remains as a vestige from the nineteenth century, when separate claim removal arose on the diversity side of the federal docket. See Edward Hartnett, A New Trick From an Old and Abused Dog: Section 1441(c) Lives and Now Permits the Remand of Federal Question Cases, 63 Fordham L. Rev. 1099, 1106-29 (1994-95). Critics of the statute abound, many of whom point out that the exercise of jurisdiction over a separate and independent claim could exceed the Gibbs definition of Article IIIs limits. See, e.g., Larry L. Tepley & Ralph U. Whitten, Civil Procedure (2d ed. 2000); Robert Casad, Personal Jurisdiction in Federal Question Cases, 70 Tex. L. Rev. 1589, 1618 (1992). Cf. Wright & Kane, supra note 12, at 235 (concluding that the statute no
1452
[Vol. 95:1423
not reach the constitutionality of exercising jurisdiction over separate but aggregated claims on the basis of minimal diversity. For one thing, Barney begins by noting that the case presents only a question of statutory construction.129 For another, the statute in question included language that authorized the federal court to remand or dismiss proceedings if it became clear at some point that the action exceeded the courts jurisdiction.130 Relying on that language, the Barney Court apparently took the position that removal of the entire case was simply a preliminary step to the eventual determination of the jurisdictional issues by the federal court: [Matters of proper joinder of parties and claims are] for the determination of the trial court, that is, the Federal Court, after the cause is there docketedif, when that is done, the cause does not really and substantially involve a dispute or controversy within the jurisdiction of that court, it can, under the fifth section of the act of 1875, dismiss the suit, or remand it to State court as justice requires.131 The Barney Court did not in terms authorize federal trial courts to hear and determine a non-diverse and separable controversy, but only permitted them to take jurisdiction of the whole cause at a preliminary stage to protect a defendants right to remove a separable controversy to federal court. Removal occurred on the basis of a separate (and diverse) controversy and might well lead to the re-alignment of the parties and some re-pleading of the claims. Once that occurred, the federal court was free to remand or dismiss the action if it exceeded federal jurisdiction. While this represented a change in procedure from the old practice of piecemeal removal, it did not necessarily portend any change in the underlying jurisdictional rule. So long as the trial court eventually dismissed or remanded the controversies as to which diversity was lacking, the remaining parties would meet the complete diversity test and no
longer serves a useful purpose and ought to be repealed). As Professor Hartnett notes, however, the exercise of removal jurisdiction over a non-federal state law claim does not necessarily violate constitutional limits. Hartnett, supra, at 1153-56 (noting that a court may exercise removal jurisdiction over non-federal claims, and then remand those that exceed the courts jurisdiction to avoid any difficulties). 129. Barney, 103 U.S. at 206. 130. See Judiciary Act of 1875, 5, 18 Stat. 470, 472 (providing that if it shall appear to the federal court, in a suit brought originally in or removed to federal court, that such suit does not really and substantially involve a dispute or controversy within the courts jurisdiction, the court shall proceed no further but shall dismiss the suit or remand it to state court). 131. Barney, 103 U.S. at 216. Earlier removal statutes called for the state court to consider whether the action was subject to removal. The 1875 Acts provision for the federal court to exercise removal jurisdiction over the entire proceeding, subject to the possible dismissal or remand of claims that exceeded federal jurisdiction, may have reflected a desire to broaden federal judicial control of the removal determination.
2007]
1453
constitutional objection could arise.132 Today, most observers assume that the lower courts should handle the removal of claims under Barneys lineal descendantsection 1441(c)in a similar fashion by dismissing or remanding any separate claims that lie beyond the scope of federal jurisdiction.133 A second commonly cited case, State Farm Mutual Ins. Co. v. Tashire, could similarly be said to stop short of endorsing minimal diversity for all purposes.134 The claims involved in Tashire grew out of a serious highway accident involving a truck and a commercial bus. Many passengers suffered injuries. But the Court refused to permit the district court to require all of the parties to join in the interpleader action; only those seeking a payment from a restricted insurance fund could be compelled to interplead their claims.135 The remaining claims were left to the ordinary processes of law. As a consequence, the Courts decision to uphold the exercise of jurisdiction over claims in the nature of interpleader represents only a modest extension of the doctrine of ancillary jurisdiction rather than a wholesale endorsement of minimal diversity. Earlier decisions had required diversity between the stakeholder and one set of claimants to the fund, and had permitted the exercise of ancillary jurisdiction over other, non-diverse claimants to the same fund.136 Tashire might be treated as
132. Practice in the federal courts in the wake of Barney confirmed that the issue of jurisdiction was to remain open throughout the litigation and could result in a remand of the non-diverse proceedings at a later stage. See Texas Transp. Co. v. Seeligson, 122 U.S. 519, 522 (1887) (following the settlement and dismissal of the separable controversy, circuit court was obliged to remand the remainder of the action to state court); Torrence v. Shedd, 144 U.S. 527, 533 (1892) (explaining that after a separable controversy has been resolved through settlement, the suit no longer really involved a dispute or controversy properly within the jurisdiction of the circuit court, and should therefore have been remanded to the state court); Prince v. Illinois C. R. Co., 98 F. 1, 3 (C.C.D. Ky. 1899) (indicating that the whole case would be removed, the separable controversy fully determined and the action returned to the state court to be there disposed of as to the other parties); Bane v. Keefer, 66 F. 610, 612 (C.C. D. Ind. 1895) (ordering remand of case following plaintiffs decision to discontinue his claims against the diverse defendant on which removal had been predicated); cf. Robinson v. Anderson, 121 U.S. 522 (1887) (ordering remand after subsequent developments revealed that a federal anchor claim lacked substance and would not support the exercise of jurisdiction). See generally John F. Dillon, Removal of Causes from State to Federal Courts (H. Black ed., 5th ed. 1889). But cf. Connell v. Smiley, 156 U.S. 335 (1895) (in action removed on the basis of a separable controversy, upholding entry of judgment in favor of plaintiff against non-diverse defendant without reaching the issue of jurisdiction). 133. See Edward Hartnett, supra note 128 (recounting the history of section 1441(c) and the dismissal or remand option). 134. See State Farm Mutual Ins. Co. v. Tashire, 386 U.S. 523 (1967). 135. See note supra. 136. Treinies v. Sunshine Mining Co., 308 U.S. 66 (1940). In focusing on the citizenship of the adverse parties, the Court followed the lead of the statute. It thus dismissed the significance of the stakeholders citizenship, treating the stakeholder as a disinterested party as between the adverse claimants and the property in dispute. See Treinies, 308 U.S. at 72. While this characterization of the stakeholder may have been accurate as to Sunshine, it was not universally true. In many cases, the stakeholder has an interest adverse to the claimants, either because they oppose interpleader or because they wish to impose liability on the stakeholder free from the demands of interpleader, or because the
1454
[Vol. 95:1423
having simply clarified that this ancillary conception of interpleader was available in an original proceeding. 2. Ancillary Jurisdiction Over True Class Actions: Supreme Tribe of Ben Hur v. Cauble The Court might also reinterpret its decision in Supreme Tribe of Ben Hur v. Cauble as resting on a theory ancillary jurisdiction.137 There, the Court treated the citizenship of the named representative of the plaintiff class as grounding the federal courts diversity of citizenship jurisdiction and ignored the citizenship of the other members of the plaintiff class.138 Cauble has long served to facilitate jurisdiction on the basis of minimal diversity: if class counsel selects a diverse representative, the presence of non-diverse class members will not destroy jurisdiction. Yet Cauble arose in a context different from that in which it has often been applied. It relied upon ancillary jurisdiction to justify the lower court in resolving the claims of non-diverse class members; it did not permit the exercise of jurisdiction over a series of separate and independent claims that had been joined together for convenient litigation.139 Of course, the Court may have some difficulty in maintaining that the ancillary context of Cauble can serve to limit the reach of that decision. But a short review of the origins of class action categories may help. Rule 23 now recognizes three different categories of class actions, which loosely map onto older categories.140 In the Rule 23(b)(1) action, class certification turns on a finding that the maintenance of separate actions could lead to inconsistent results and incompatible standards of conduct or would impair
stakeholder denies all liability as an alternative ground. See Zechariah Chafee, Jr., Interpleader in the United States Courts, 41 Yale L.J. 1134, 1141-43 (1931) (discussing the alignment of opposing interests at the first and second stages of interpleader and discussing the implications of such alignments for the existence of diversity and ancillary jurisdiction). 137. 255 U.S. 356 (1921). The lower courts have interpreted Cauble to mean that only the citizenship of the named representative of a class counts for purposes of determining the citizenship of the class for diversity purposes. See, e.g., Aetna Cas. & Surety Co. v. Iso-Tex, Inc., 75 F.3d 216 (5th Cir. 1996); In re Agent Orange Product Liability Litigation, 818 F.2d 145, 162 (2d Cir. 1987); In re School Asbestos Litigation, 921 F.2d 1310, 1317 (3d Cir. 1990). 138. See, e.g., Aetna Cas. & Surety Co. v. Iso-Tex, Inc., 75 F.3d 216 (5th Cir. 1996); In re Agent Orange Product Liability Litigation, 818 F.2d 145, 162 (2d Cir. 1987); In re School Asbestos Litigation, 921 F.2d 1310, 1317 (3d Cir. 1990). 139. Ancillary jurisdiction is discussed supra at pages 16-17. Ancillary jurisdiction extends to claims that bear such a relationship to a controversy properly before the court that they cannot, in fairness to an interested party, be excluded from the litigation unit. 140. See Fed. R. Civ. P. 23. Rule 23 was amended in 1966, to create the familiar (b)(1), (b)(2), and (b)(3) categories that it now contains. Before that, Rule 23 provided for the certification of true, hybrid and spurious class actions, actions that roughly mapped onto the new categories. For an account of the 1966 amendments, and the concerns with the old categories that led to the change, see Robert G. Bone, Personal and Impersonal Litigative Forms: Reconceiving the History of Adjudicative Representation, 70 B.U. L. Rev. 213, 291-304 (1990). On the history of the class action as an equitable device, see Stephen C. Yeazell, From Medieval Group Litigation to the Modern Class Action (1987).
2007]
1455
or impede an absentee partys ability to protect her interests. In a (b)(2) action, class certification follows a finding that the defendant has acted on grounds generally applicable to a class thereby making injunctive relief proper as to the class as a whole. In a (b)(3) action, certification depends on a showing that the claims of class members present common questions of fact or law which, when viewed in light of other factors, can best be handled through class treatment. These three categories of class actions roughly correspond to what were once termed true, hybrid and spurious class actions, each with its special rules. The differing theoretical justification for class treatment across categories gave rise to differences in the scope of jurisdiction. The dispute in Cauble arose from the adjudication of a true class action, one that courts would today characterize as a (b)(1) class action.141 In such proceedings, the complete diversity rule did not present a problem. Rather, the Cauble Court explained that the principle of ancillary jurisdiction controlled the power of the district court to exercise jurisdiction over the claims of non-diverse class members.142 As a result, it was not necessary to join the non-diverse members as parties to bind them to the decree; like creditors in an equity receivership, their citizenship was simply irrelevant once the district court obtained jurisdiction over the proceeding.143 The risk
141. The class litigation began when one Balme, a citizen of Kentucky and a member of a beneficial insurance society, brought suit in federal district court against the society and its officers (all Indiana citizens) to enjoin the implementation of a new set of rules that were to alter the members insurance benefits. The plaintiffs proceeded on behalf of a class of several thousand beneficiaries who were located around the country, but made no effort to notify or include the various members from Indiana. The district court entered judgment in favor of the society, thus rejecting the challenge to the new rules. Later, citizens of Indiana brought similar suits against the society in Indiana state court, mounting essentially the same challenge. The society returned to federal court for an injunction against the prosecution of these state court actions, contending that all members of the society had been members of the class and were precluded by the prior litigation. The issue of preclusion depended on whether the federal court had adjudicated the claims of the absentee Indiana class members in the first proceeding. The Indiana class members argued that, as citizens of the same state as the defendants, the district courts diversity jurisdiction did not extend to their claims. 142. In doing so, the Cauble Court relied on its earlier decision in Stewart v. Dunham, 115 U.S. 61 (1885) There, a creditor initiated an equitable proceeding to set aside a fraudulent transfer. Following removal on the basis of diversity, additional plaintiffs joined the proceeding. While their citizenship would have destroyed diversity, the Court concluded that their claims were ancillary to the jurisdiction acquired between the original parties. Id. at 64. This ancillary jurisdiction was said to extend to the claims of any creditors who were entitled to benefit from a decree setting aside the fraudulent transfer. As the Stewart Court explained, once the decree issued in a diverse proceeding, a special master would have had authority to administer the decree by allowing proof of claims against the assets fraudulently transferred. Id. 143. Balme, the Kentucky citizen, was said to have a right to pursue claims against the society and its officers in federal court on the basis of diversity, much the way a diverse shareholder could bring a derivative action on behalf of a class of shareholders against the corporation in which she owned stock. Once Balmes claim was properly before the federal court, the court was said to enjoy ancillary jurisdiction over the claims of all the members of the same class. Otherwise, the society would face the risk of conflicting or inconsistent judgments, some of which might uphold the rule changes while others invalidated them. Since the members were all in the same position, their rights and liabilities [were all]
1456
[Vol. 95:1423
of incompatible results and the obvious commonality of claims were seen as requiring resolution in a single proceeding.144 Limited to recognizing ancillary jurisdiction over true class actions, Cauble had no occasion to discuss the jurisdictional rules that applied to other categories of class litigation. Spurious class actions, in particular, were poor candidates for application of the Cauble approach. In spurious proceedings, the class action operated as a joinder device for multiple parties.145 Such proceedings bound only those who actually appeared before the court. (The modern (b)(3) class action shifts gears somewhat by requiring members of the class to opt out after receiving notice if they do not wish to participate; spurious class members were participants only if they opted in.) Such spurious proceedings presented no risk of incompatible determinations and no justification for the exercise of ancillary jurisdiction. Spurious class actions do not seek injunctive relief and do not confront the defendant with the threat of incompatible standards of conduct. Nor will the adjudication of claims in separate proceedings ordinarily impede other parties ability to protect their interests; only where a limited fund exists of the kind that brings into play interpleader-style justifications for joinder will any impediment arise and that, of course, would create a true class action proceeding.146 To date, the Court has not seen fit to consider the application of Cauble to (b)(3) class actions or to consider why the lower courts have extended the Cauble rule.147 Rather, it has simply accepted the citizenship
before the court, especially where the subject matter of the suit was common to all. Cauble, 255 U.S. at 363. This was not a case, in short, of separate and distinct claims, joined for convenience or on the basis of transactional relationship, but was a true class action that brought into play principles of ancillary jurisdiction. 144. Supreme Tribe of Ben Hur v. Cauble, 255 U.S. 356, 366-67 (1921). 145. See Steel v. Guaranty Trust Co., 164 F.2d 387, 388 (2d Cir. 1947) (describing the spurious class action as in effect, but a congeries of separate suits so that each claimant must, as to his own claim, meet the jurisdictional requirement); see also Hackner v. Guaranty Trust Co., 117 F.2d 95 (2d Cir. 1941). 146. Lower court decisions understood the ancillary nature of the jurisdiction at issue in Cauble and the inapplicability of that basis for jurisdiction in cases involving separate claims joined in a spurious class action. Thus, lower courts were willing, in true class actions to exercise ancillary jurisdiction over unnamed class members and to aggregate class claims to satisfy the amount-incontroversy threshold. See Calagaz v. Calhoun, 309 F.2d 248 (5th Cir. 1962) (citizenship of the named representative controls the citizenship inquiry in a true class action); Rosenberg v. Chicago Title & Trust Co., 128 F.2d 245 (7th Cir. 1942) (true class action puts into issue the entire value of the trust or fund, without any need to aggregate separate claims). But the lower courts were unwilling to permit aggregation in spurious class actions. See, e.g., Troup v. McCart, 238 F.2d 289 (5th Cir. 1957); Giordano v. Radio Corp. of America, 183 F.2d 558 (3d Cir. 1950); see also Zahn v. International Paper Co., 414 U.S. 291, 297 n.7 (collecting lower court authority). 147. In Snyder v. Harris, the Court expressed concern that an aggregation approach might seriously undercut the congressional policy of using a statutory amount to restrict access to the diversity dockets of the federal courts. As part of its expression of policy concerns, the majority opinion of Justice Black described the docket-expanding potential of the Cauble rule as if it applied to all class actions; the opinion failed to distinguish between the spurious or (b)(3) actions at issue in Snyder and
2007]
1457
determinations of the lower courts without reaching the issue. For example, in its most recent decision on jurisdiction over class actions, Exxon Mobil Co. v. Allapattah Services Inc., the Court interpreted section 1367 as overruling Zahn v. International Paper Co.148 and thus permitting district courts to exercise supplemental jurisdiction over claims of the members of (b)(3) class actions that fail to meet the amount-in-controversy threshold.149 But other portions of its opinion, albeit in dicta, took pains to preserve the complete diversity requirement in the face of an argument that the supplemental jurisdiction statute had unwittingly overturned that longstanding rule as well.150 While the lower court presumably applied Cauble in determining the citizenship of the plaintiff class,151 the Exxon
the true class action that, in Cauble itself, had been seen as justifying the exercise of ancillary jurisdiction. See Snyder v. Harris, 394 U.S. 332, 340 (1969) (citing Cauble for the proposition that [u]nder current doctrine, if one member of a class if of diverse citizenship from the class opponent, and no nondiverse members are named parties, the suit may be brought in federal court even though all other members of the class are citizens of the same State as the defendant and have nothing to fear from . . . the courts of their own State.) Lower court opinions seized upon the Snyder dicta in applying the Cauble rule to (b)(3) class actions, apparently failing to recognize Justice Blacks error. See In re Agent Orange Product Liability Litigation, 818 F.2d 145, 162 (2d Cir. 1987) (citing Justice Blacks dictum in Snyder to justify application of the Cauble rule to (b)(3) class action); see also In re School Asbestos Litigation, 921 F.2d 1310, 1317 (3d Cir. 1990) (same). Although a few lower court judges have leaned against the trend, see Rosmer v. Pfizer, Inc., 272 F.3d 243, 251 n.2 (4th Cir. 2001) (Niemeyer, J.) (dissenting from denial of rehearing en banc; arguing that the Cauble rule should not apply to (b)(3) class actions); see also Dudley O. McGovney, A Supreme Court Fiction II, 56 Harv. L. Rev. 1090, 1103-15 (1943), most observers simply accept the dominant view as to the breadth of the Cauble rule. See Wright & Kane, supra note 12, at 187. 148. Zahn v. International Paper Co., 414 U.S. 291 (1973). Zahn refused to permit the assertion of supplemental jurisdiction over the members of a plaintiff class whose claims failed to meet the amount in controversy. 149. Exxon Mobil Co. v. Allapattah Servs. Inc. 545 U.S. 546, 558 (2005) (The single question before us, therefore, is whether a diversity case in which the claims of some plaintiffs satisfy the amount-in-controversy requirement, but the claims of other plaintiffs do not, presents a civil action of which the district courts have original jurisdiction.). 150. See Id. at 549, 559, 562 (assuming the existence of diversity of citizenship and focusing entirely on the application of the amount-in-controversy requirement; noting that the complete diversity rule would foreclose original jurisdiction if non-diverse parties appeared on both sides of the action even in the face of language in the supplemental jurisdiction statute that could be literally read to the contrary). Both Justice Ginsburg and academic commentators have questioned the majoritys decision to treat the complete diversity and amount-in-controversy rules differently for jurisdictional purposes. See Id. at 558 (Ginsburg, J., dissenting); Adam N. Steinman, Sausage-Making, Pigs Ears, and Congressional Expansion of Federal Jurisdiction: Exxon Mobil v. Allapattah and its Lessons for the Class Action Fairness Act, 81 Wash. L. Rev. 279 (2006). 151. Neither the district court nor the appellate court opinion expressly relies on Cauble. But the principal defendant at the time of the suits initiation, Exxon Corporation, was a citizen of New Jersey, its state of incorporation, and of Texas, its principal place of business. See Allapattah Servs. Inc. v. Exxon Corp., 333 F.3d 1248 (11th Cir. 2003), affd, 545 U.S. 546 (2005). The class included 10,000 dealers, some of whom were located in the state of Texas. Thus, consideration of the citizenship of all members of the plaintiff class would have seemingly destroyed diversity.
1458
[Vol. 95:1423
Mobil Court did not discuss or approve the Cauble rule and did not review that aspect of the lower courts decision.152 In sum, the Courts relaxation of the strict version of the complete diversity rule does not imply the absence of all limits; its leading cases allow the removal of separate claims or the assertion of ancillary jurisdiction over the claims of parties needed for a just adjudication. Cognizant of this fact, early proposals to relax the complete diversity requirement for certain forms of complex, multi-party litigation were careful to identify ancillary jurisdiction as the predicate for doing so. Thus, the 1969 ALI Study on the Division of Jurisdiction Between State and Federal Courts proposed to add a new section to the judicial code that would have authorized jurisdiction over multi-party litigation involving diversity between any two adverse parties.153 But as Professor Floyd observes, the ALI provision authorized the exercise of such minimal diversity jurisdiction only where joinder of additional defendants was necessary for a just adjudication of the plaintiffs claim.154 The ALIs commentary on this multi-party provision underscored the ancillary character of the jurisdiction under contemplation; the necessary-party requirement was meant to express a degree of urgency for the presence of scattered parties which goes beyond trial efficiency and economy.155 Tashires reliance on a memorandum in the ALI Study, approving an ancillary-based form of minimal diversity, does not necessarily compel the acceptance of the broader forms of minimal diversity jurisdiction reflected in recent legislation.156 As a consequence, none of the leading minimal
152. Writing for the majority, Justice Kennedy assumed that the claims of the absent class members form a part of the same case or controversy as the claims of the named representatives. Exxon Mobil, 545 U.S. at 558. At another point, he observed that the single issue before the Court is whether a diversity case in which the claims of some of the plaintiffs satisfy the amount-in-controversy requirement, but the claims of other plaintiffs do not, presents a civil action of which the district courts have original jurisdiction. Id. at 556. The decision thus consciously avoids the issue of diversity of citizenship and the proper breadth of an Article III controversy, and focuses instead on the statutory amount-in-controversy requirement. Indeed, at one point, the Court indicated that the absence of complete diversity operates as a more potent bar to jurisdiction than the absence of the requisite amount in controversy. Id. at 562. This defense of the complete diversity rule looks a bit odd when juxtaposed against its refusal to look beneath the lower courts application of the Cauble rule. 153. ALI Study, supra note 101, at 2371(a). 154. See Floyd, supra note 12, at 655 (quoting ALI Study, supra note 101, at 2371(a)). Section 2371(b) went on to define necessary parties in terms of parties needed for a just adjudication. This would have permitted the assertion of jurisdiction over indispensable parties, but not over joint tortfeasors or others on whom the plaintiff sought to impose joint and several liability. See ALI Study supra note 101, at 68. 155. See Floyd, supra note 12, at 655 (quoting id. at 385). See also ALI Study, supra note 101, at 431 n.14 (observing that the ALI multi-party provision could be fully supported on an ancillary jurisdiction theory). 156. Thus, the reporters memorandum in support of the availability of minimal diversity identifies ancillary jurisdiction as a decisive argument against the complete diversity rule of Strawbridge v. Curtiss. See Floyd, supra note 12, at 659 n.200 (quoting ALI Study, supra note 101, at 434, which concluded, after an analysis of Strawbridge, that the concepts underlying ancillary
2007]
1459
diversity casesBarney, Ben Hur and Tashireresolve the question of jurisdiction over separate claims aggregated for convenience in a (b)(3) class action. B. CAFA and Rule 23(b)(3) Class Actions
In evaluating the constitutionality of CAFAs jurisdictional provisions, the Court might begin by recognizing that CAFA challenges the assumptions that normally govern the allocation of substantive and procedural lawmaking in our federal system.157 CAFA does not regulate the substantive rights of individual consumers or firms: the claims that comprise class actions under CAFA arise under state law, and do not include any substantial federal ingredients of the kind that might support the exercise of federal question jurisdiction. CAFA relies on jurisdictional expansion to secure the application of an existing set of federal procedural rules so as to regulate state court class action practice.158 To be sure, one can argue that Rule 23 and its state law analogs have made important and effectively substantive changes in the rights of litigants by allowing the aggregation of the small or negative value claims of individual litigants that would not otherwise warrant litigation.159 But Rule 23 was designed to apply to claims otherwise properly before the federal courts; it was not designed to provide a substantive federal right to aggregation for class
jurisdiction would authorize Congress to confer jurisdiction with less than total diversity). For the ALI, at least in the 1960s, the justification for minimal diversity lies not in the efficiency and economy associated with jurisdiction over related but separate claims but in avoiding the extra burdens that litigation of a diverse-party controversy would impose on those needed for the just adjudication of a controversy that has already been brought within federal jurisdiction. See ALI Study, supra note 101, at 434. 157. At least since the Erie decision, state courts and legislatures have borne primary responsibility for the development of state law, even as federal courts occasionally hear state law questions in the exercise of diversity or supplemental jurisdiction. By the same token, the primary responsibility for creating and applying federal substantive law falls to Congress and the federal courts, respectively, with the ultimate interpretive authority lodged in the Supreme Court. Federal procedural rules play a supporting role in the enforcement of rights grounded in state and federal substantive law, operating to facilitate the fair and efficient resolution of claims before the federal courts. Under the Rules Enabling Act, such procedural rules may not abridge, modify, or enlarge substantive rights, see 28 U.S.C. 2072, and the Court has strained in recent years to narrow the operation of the Federal Rules to prevent their interfering with substantive interests, both state and federal. See infra notes 160. 158. CAFA does include a modest collection of rules to govern the class certification and settlement process, but these do not differ in the main from Rule 23s approach and do not apply in state court. 159. See Edward H. Cooper, Rule 23: Challenges to the Rulemaking Process, 71 N.Y.U.L. Rev. 13 (1996) (noting that Rule 23s nominally procedural approach to aggregation had nonetheless made an important change in the nature of the underlying rights, transforming low or negative value claims into claims worth pursuing, and reflecting on the challenges such developments pose to ongoing procedural reform); David L. Shapiro, Class Actions: The Class as Party and Client, 73 Notre Dame L. Rev. 913 (1997) (adopting the view that class actions, especially those involving the aggregation of low-value claims, may be best regarded as an entity and exploring the implications of such a view).
1460
[Vol. 95:1423
actions based upon state substantive law.160 If the Court accepted the premise that procedural rules cannot provide the substantial federal ingredient needed to ground jurisdiction under Osborn, it might well refuse to permit the assertion of federal question jurisdiction over the aggregated claims in question. The Court might also conclude that CAFA brings controversies before the district court that lack the constitutionally-required diversity of citizenship. Rule 23(b)(3) class actions propose to aggregate a series of separate controversiesoften the claims of a firms customers for breach of contract or violation of a consumer protection statutethat share questions of law or fact in common. CAFA provides for jurisdiction over this bundle of separate controversies, so long as any one of them satisfies the complete diversity requirement. Note that Congress has not expanded jurisdiction over individual claims on the basis of diversity of citizenship. Individual claims must still satisfy the complete diversity and amount-in160. Federal courts apply federal procedural rules in disputes over both state and federal law, but the rules shall not be construed to extend the jurisdiction of the district courts. Fed. R. Civ. P. 82. Moreover, under the Rules Enabling Act, any rules promulgated by the Court shall not abridge, enlarge, or modify any substantive right. 28 U.S.C. 2072(b). Views differ as to the purpose of this restriction on the Courts rulemaking authority. Compare Stephen B. Burbank, The Rules Enabling Act of 1934, 130 U. Pa. L. Rev. 215 (1982) (portraying the limitations in the Act as aimed at preserving Congresss control) with John Hart Ely, The Irrepressible Myth of Erie, 87 Harv. L. Rev. 693 (1974) (exploring the Acts restrictions as protecting the state law from displacement by rules of federal procedure). But whatever ones view, the Act plainly forecloses an exercise of rulemaking authority that would create a federal substantive right to aggregation under which state law claims could arise for jurisdictional purposes. See Shapiro, supra note 159, at 949-54 (concluding that the adoption of an entity theory of the class action would implicate matters of substance that defy reliance on federal common law solutions and supporting the development of the necessary rules by act of Congress rather than by rulemaking). One can certainly imagine an argument that the federal courts should decline to apply Rule 23 standards under CAFA in deference to state rules that point to a different decision about the propriety of aggregate litigation. Imagine a state law that forbids the aggregation of certain kinds of claims, or permits their aggregate treatment in circumstances in which Rule 23 would not. Such differences in approach have already created forum shopping pressures, with outcomes turning on what aggregation standard applies. See S. Rep. No. 109-14, supra note 108, at 26 (noting the tendency of defendants to remove actions to federal court where the threat of prejudice is significantly lower). When facing such a conflict, the federal courts may consider a narrowing interpretation of Rule 23 that prevents the federal aggregation rule from overriding state rules with substantive overtones, at least as to claims based upon state law. See Gasperini v. Center for Humanities, Inc., 518 U.S. 415 (1996) (adopting a narrow interpretation of federal procedural rules to avoid a conflict that would have otherwise displaced the applicability of a more search state law standard for the review of jury verdicts); see also Walker v. Armco Steel Corp., 446 U.S. 740 (1980) (adopting a narrow interpretation of the federal commencement rule to prevent a displacing conflict with state rules governing the tolling of the limitations period). For a criticism of this trend toward selective narrowing of the federal rules, see Earl C. Dudley, Jr. & George Rutherglen, Deforming the Federal Rules: An Essay on Whats Wrong with the Recent Erie Decisions, 92 Va. L. Rev. 707 (2006). If such narrowing occurred, it would seemingly undermine the congressional goal of securing the application of Rule 23 standards to all multi-state class actions and would substitute a patchwork quilt of certification standards. Even if such narrowing were avoided, the price of shifting to a federal Rule 23 standard would be the displacement of state rules with partially substantive features.
2007]
1461
controversy requirements under the familiar provisions of section 1332(a).161 Only when the plaintiffs aggregate individual claims in a class action do CAFAs jurisdictional provisions come into play. The statutes focus on the aggregation of individual claims could provide the Court with a plausible basis for distinguishing traditional notions of ancillary jurisdiction. CAFA does not seek to justify the jurisdictional expansion to include claims that might be necessary for the just adjudication of the claims by plaintiffs properly before the district court on diversity grounds. CAFA focuses on the class as a whole, and not on the claims of individual plaintiffs. Fairness to defendants might offer a better case for the exercise of ancillary jurisdiction. Congress enacted CAFA in part because state court class action practices unfairly burden corporate defendants with overlapping and duplicative litigation. One can link these fairness concerns to the original justification for ancillary jurisdiction, which rested on the view that once the court had acquired jurisdiction over property or a limited fund, its ancillary jurisdiction extended to other parties with a legal interest in that property or fund.162 The threat of multiple litigation that grounds ancillary jurisdiction over non-diverse claimants in an interpleader action bears some resemblance to the problem of overlapping class action jurisdiction. The connection between overlapping class actions and the notion that defendants have an interest in their consolidated resolution provides an important part of CAFAs justification. But the Court might defensibly respond that ancillary jurisdiction comes into play only after a dispute between diverse citizens has been framed; it is the aggregation of claims into a class action, rather than the individual claim of a single diverse member of a plaintiff class, that creates the problem of duplication. The emphasis on aggregation in questioning the availability of traditional notions of ancillary jurisdiction may force the Court to confront a proposal to reconceptualize CAFA. Perhaps CAFA transforms the interstate class action from an aggregation of separate controversies into a single joint controversy. Perhaps any diversity of citizenship among the parties that comprise the joint interests of the class should then suffice to establish diversity, just as with the case of the true class action recognized in Cauble. Recent theorizing about the nature of class litigation may offer some support for the joint or entity-based treatment of the claims that comprise a (b)(3) class action. With his customary grace, Professor Shapiro has suggested that class actions may be best understood as entities with a juristic status separate from their members, rather than as aggregations of
161. When they do, supplemental jurisdiction over related claims may attach under the Exxon Mobil Courts interpretation of section 1367, thereby moderating the amount-in-controversy but not the complete diversity requirement. 162. See supra notes 103-110 and accompanying text.
1462
[Vol. 95:1423
individual claimants.163 Building on Professor Shapiros work, Professor Issacharoff has explored the implications of such an entity theory for the application of rules of procedural due process to the members of small or negative value (b)(3) class actions.164 Although Professor Redish has raised important due process questions about the aggregation of the individual claims that make up a class action,165 the work of Shapiro and Issacharoff rightly notes that something important changes when rulemakers approve aggregate litigation. How should the Court address the claim that Congress in CAFA regulated the class action as an entity, rather than providing for the exercise of jurisdiction over an aggregate collection of individual claims?166 The Court might respond by noting the important differences between the true classes that give rise to joint interests under Cauble and the spurious classes that arise through a (b)(3) aggregation.167 Modern juristic
163. See Shapiro, supra note 159. 164. See Samuel Issacharoff, Preclusion, Due Process, and the Right to Opt Out of Class Actions, 77 Notre Dame L. Rev. 1057 (2002). 165. See Martin H. Redish & Nathan D. Larsen, Class Actions, Litigant Autonomy, and the Foundations of Procedural Due Process, 95 Calif. L. Rev. 1587-1600 (2007) (criticizing entity theories on the ground that they accord too little weight to the individuals interest in controlling his or her own legal claims). 166. One can usefully contrast CAFA with Congresss earlier efforts to expand jurisdiction over juristic entities. As Congress saw the problem that led it to adopt the Labor Management Relations Act of 1947, unions escaped liability for the breach of their labor contracts due in part to the absence of any federal duty to observe them and in part to the failure of state courts to develop the rules of aggregate litigation necessary to enforce such promises against the union. Many states viewed the union as an unincorporated association, incapable of entering into contractual obligations as an entity and suable only through the joinder of its members. State court suits to remedy strikes in breach of a labor contract presented both of these aggregation problems. Congress addressed the problem by creating certain federal substantive rights and securing their enforcement through rules that conferred entity status on the union. In section 301(a) of the LMRA, it provided for the federal courts to assert jurisdiction over suits for violation of collective bargaining agreements. (Other provisions of the new law imposed liability on unions for secondary boycotts.) In section 301(b), Congress established rules of litigation capacity. In brief, those rules make the union an entity for purposes of enforcing the unions obligations under federal law, giving the union capacity to sue and be sued in its own name. To secure this new entity status, the union was made responsible for actions taken on its behalf by its officers, but the officers and members were freed from any personal liability for the obligations of the union (thus providing something analogous to the corporations limited liability). Actions brought against the union under federal law could name the union itself, without the cumbersome necessity of joining the members through some form of aggregate litigation. On the operation of 301(b), and its tendency to confirm Congresss intent to create a federal duty to honor collective bargaining agreements, see Pfander, supra note 38, at 292-98. 167. Apart from litigation to enforce federal substantive rights, unions may sue and be sued in federal diversity proceedings to enforce garden-variety state law tort and contract claims. Rule 23.2 sets forth a special class action rule that empowers an officer of the union to pursue state law claims on the unions behalf in federal court, so long as the representative party and the defendant are citizens of different states. This approach to defining the citizenship of the union as a class flows from the Cauble decision and its emphasis on the citizenship of the representative plaintiff in determining the citizenship of a true class. This true class characterization seem quite apt in the case of a labor organization: Perhaps to a greater extent than in the case of the shareholders of a corporation, the members of a labor organization share a community of interest as a result of their membership status.
2007]
1463
entities, like corporations and labor unions, adopt constitutions and bylaws to structure their internal affairs. These foundational documents define the rights and obligations of membership, establish a board of directors, and assign certain officers responsibility for the day-to-day operation of the union. Decisions about litigation naturally fall to the unions officials. The members of a (b)(3) class, by contrast, often have no relationship other than that created when the lawyer for one class member proposes to join their several claims together for aggregate litigation. The absence of any pre-existing relationship among members creates many of the pitfalls that we now associate with (b)(3) class litigation. Unlike corporations and unions that vest control in officers, no particular member of a (b)(3) action can claim to have been identified by the group as the representative of all the members. This lack of any identified class representative leads to the possibility that a single volunteer, usually a lawyer, will initiate the litigation. Other class members can pursue the same claims in other forums, perhaps with different lawyers at the helm, an approach that juristic entities would rarely adopt. In many cases, no single member of the class has adequate incentives to monitor the progress of the litigation and the class lacks any internal system of organization that would enable the members to make informed decisions about who, if anyone, should pursue claims on their behalf.168 These observations reveal that Congress has ample power to fashion a true class or juristic entity or to take the steps necessary to bring class actions within the federal question or ancillary jurisdiction of the federal courts. If Congress chooses to create a new federal entity and regulate its internal affairs, it can presumably enable that entity to sue and be sued within the Osborn federal ingredient tradition. But as Justice Frankfurter
Conflicts may ensue, especially among the unions leaders and their political opponents. But the members share a common interest in securing compensation for the union from any outside entity that breached its contract with the union. In this sense, the unions contract claims represent a common and undivided interest of all the members, rather than separate and independent claims. The rules of ancillary jurisdiction permit litigation through a class action and the selection of a single diverse member as the representative of all. It might make more sense if Congress were to abandon the Rule 23.2 approach and draw the analogy to the corporation more explicitly. It could so by defining the labor organization, and other unincorporated associations, as citizens of their place of organization and perhaps their principal place of business. The advantage of such an approach would be that it would fix in advance the citizenship of the labor union, rather than permitting the union or its opponent to manipulate its citizenship through the selection of a representative member whose citizenship differs from that of the adverse party. Such an approach would also help to assure adequacy of representation in litigation on behalf of the union by making clear that the union itself would control litigation decisions through its duly elected officers. Rule 23.2 class actions might remain for use in a members derivative action, something comparable to the shareholders derivative action envisioned in Rule 23.1. 168. These observations should not be taken as rejecting the entity model of class litigation that some scholars have proposed. Rather, the point is to recognize that the adoption of an entity model will require Congress to adopt a law that defines the rules of entity creation that it envisions as necessary to protect the interests of class members and perhaps specify internal rules of governance to address potential conflicts of interest.
1464
[Vol. 95:1423
noted in Lincoln Mills, the congressional adoption of merely procedural rules should not suffice to ground the exercise of Osborn-style jurisdiction over the claims of the entity.169 As applied to the multi-state class actions embraced by CAFA, then, the Osborn approach would enable Congress to define those situations in which it deems aggregate litigation by a federal entity appropriate, so long as it specifies the rules that govern the entitys creation and internal operation. Like federal corporations that Congress has empowered to sue and be sued in federal court on state law causes of action, federal litigation entities might well be given access to federal court on the theory that their federal creation provides an original ingredient in every action they bring. Entity status, if congressionally conferred, might provide a broad foundation for federal question jurisdiction. Even if these new entities were the creatures of state law (such as most business corporations and many unincorporated associations), their members might constitute a true class that, under Cauble, would justify an assertion of ancillary jurisdiction on the basis of minimal diversity. Yet the Court might conclude that (b)(3) class actions under CAFA lack the elements that would justify either federal question or ancillary jurisdiction. Congress has not written rules to bring into existence new federal litigation entities of the kind that might trigger Osborn-style jurisdiction. Moreover, the members of a (b)(3) class do not engage in the voluntary joinder decisions that are now the hallmark of the entities that qualify for true class treatment. Most true class members make a voluntary decision to associate themselves with others in the common enterprise. Shareholders purchase shares of stock in the firm and can typically leave the enterprise by selling their shares later. Members of labor unions accept employment in a bargaining unit represented by a union or join their fellow workers in forming a union.170 While the (b)(3) class action provides a
169. Textile Workers Union v. Lincoln Mills, 353 U.S. 448, 460 (1957) (Frankfurter, J., dissenting) Congress, perhaps as early as 1947 and certainly by 1959, had transformed the labor union from a voluntary organization under state law to the functional equivalent of a corporate entity, governed by a federal law of internal affairs that resembles the Delaware corporation code. Having created in some respects a federal entity, Congress would seemingly have had the power under Osborn v. Bank of the United States to confer jurisdiction on the federal courts to entertain claims by and against the labor union as claims arising under federal law. Justice Frankfurter expressed doubt on this point in his Lincoln Mills dissent, characterizing the rules of entity treatment and union capacity as simple matters of procedure that could not provide the sort of federal substance necessary to supply an original ingredient in any action the union might bring. But the passage of the LMRDA unquestionably constitutes substantive federal regulation of internal affairs that makes it impossible to sustain Justice Frankfurters procedural characterization of the unions status as a federal entity. Even in 1947, there were rules of substance lurking in the details of the entity status of labor unions that Justice Frankfurter characterized as merely procedural. See Pfander, supra note 38, at 292-98. 170. Federal law protects the right of employees to refrain from formal union membership, if they choose, even as it permits the union to charge such objectors a ratable share of the unions costs in
2007]
1465
formal opportunity for prospective class members to opt out of the proceeding, many observers view that decision as something less than a voluntary association with the class. Class members could rationally decline to take any action at all on receipt of a notice of class litigation, a fact that suggests to many that their failure to opt out may not signal an affirmative desire to participate in the litigation. For this reason, it is difficult to characterize the (b)(3) class as a voluntary association of likeminded claimants whose common interests justify true class treatment and the corresponding exercise of ancillary jurisdiction. If the Court were inclined to conclude that conceptions of the (b)(3) class as an entity fail to provide a basis for upholding the jurisdictional grants in CAFA, the Court might draw upon the new skepticism about the aggregation of individual claims that animates much current thinking about class actions. Critics of the settlement class action, notably including the Court itself, have emphasized the importance of attending to the substantive rights of individual claimants.171 Scholars concerned with procedural due process have resisted class action solutions that accord too little weight to the interests of claimants in a fair resolution of their individual claim.172 Recent decisions emphasize the right of individuals to a trial on issues of damages in resisting the shift away from the notice requirements in (b)(3) actions.173 The Court might invoke these doubts in suggesting that CAFAs reliance on the aggregation of separate claims exceeds the limits of Article III. C. Federal Substance and Aggregate Jurisdiction
If the Court identified the absence of a federal substantive right to aggregate litigation as a plausible basis for doubting the jurisdictional foundation of CAFA, it might also question Congresss recent decisions to
representing the bargaining unit. Ultimately, objectors may leave the unionized workplace and take up work in the non-union sector. 171. See Amchem Prods. Inc. v. Windsor, 521 U.S. 591 (1997); Ortiz v. Fibreboard, 527 U.S. 815 (1999) (invaliding settlement class actions that failed to give adequate attention to the conflicting interests of the many claimants encompassed within the settlement; expressing skepticism about the use of the class action as a device to alter the substantive rights of individual litigants in pursuit of the goal of convenience and efficiency). See also Paul D. Carrington & Derke P. Apanovitch, The Constitutional Limits of Judicial Rulemaking: The Illegitimacy of Mass-Tort Settlements Negotiated Under Federal Rule 23, 39 Ariz. L. Rev. 461 (1997) (expressing skepticism about class action settlements that subordinate individual rights to collective interests in efficient disposition); Nagareda, supra note 110, at 677 (arguing that class actions settlements cannot provide a national solution to mass litigation and cannot substitute for national lawmaking). 172. See Redish & Larsen, supra note 165, at 1600-1612; Susan P. Koniak & George M. Cohen, Under Cloak of Settlement, 82 Va. L. Rev. 1051 (1996). 173. See Alison v. Citgo Petroleum Corp., 151 F.3d 402, 411 (5th Cir. 1998) (limiting use of a (b)(2) class action when individual claims for money damages predominate); Jefferson v. Ingersoll Intl, 195 F. 3d 894, 898 (7th Cir. 1999); but cf. Robinson v. Metro North R. Co., 267 F. 3d 147, 165 (2d. Cir. 2001).
1466
[Vol. 95:1423
expand federal jurisdiction over certain mass disasters. Under the Multiparty Multiforum Trial Jurisdiction Act (MMTJA), Congress authorized the assertion of jurisdiction over litigation that stems from a single accident at a discrete location that claims the lives of at least seventy-five persons. When the requisite triggering accident has occurred, the statute authorizes the district courts to exercise original and removal jurisdiction on the basis of minimal diversity between adverse claimants. The two most expansive provisions are those that permit a non-diverse plaintiff to intervene in any pending district court proceeding involving claims arising from the same accident that satisfy the minimal diversity test and those that permit defendants to remove a non-diverse state proceeding to federal court so long as the defendant has become a party to other litigation that does satisfy minimal diversity. These provisions authorize the assertion of diversity jurisdiction over claims between non-diverse parties so long as the claims arise from a qualifying accident that produces litigation otherwise meeting the minimal diversity test. Although the MMTJA provides for jurisdiction over separate controversies between co-citizens, one might defend the statute on federal question grounds. The federal jurisdictional trigger-the death of seventyfive persons at a discrete location-might arguably provide an original ingredient that would justify an assertion of federal question jurisdiction over these separate claims. Like the jurisdictional trigger that the Court in Verlinden viewed as sufficient to justify federal question jurisdiction over claims under the Foreign Sovereign Immunities Act (FSIA),174 the seventyfive-death trigger under the MMTJA might be viewed as a federal law standard inherent in every claim arising under the Act. As with the FSIA, one might view the MMTJA as authorizing the adjudication of claims based upon state law only when the triggering question of federal law has first been resolved. Yet the federal law trigger of jurisdiction under the MMTJA does not appear sufficiently substantive to warrant federal question jurisdiction. Note that unlike the FSIA, where liability and jurisdiction turn on a finding that the acts of the foreign sovereign fall within the Acts various commercial activity exceptions to immunity, liability in a mass disaster case does not turn on the MMTJAs trigger of jurisdiction. Liability remains a creature of state law and the liability rules that apply will be the same whether the mass disaster claims the lives of 74 or 75 individuals. As with CAFA, then, the trigger of jurisdiction under the MMTJA does not alter substantive rights but serves to facilitate federal procedural coordination of claims based on state law. One might usefully contrast the jurisdictional provisions of the MMTJA with those of another well-known mass disaster statute, the Air
174. See Verlinden B. V. v. Central Bank of Nigeria, 461 U.S. 480, 496-97 (1983).
2007]
1467
Transportation Safety and System Stabilization Act (ATSSSA), which Congress adopted in the wake of the September 11 attacks.175 By its terms, ATSSSA creates a federal cause of action to recover damages for injuries suffered in the four airline crashes that occurred on that date. (Note that the MMTJA was not yet on the books.) Moreover, the Act declares that the action shall be the exclusive remedy for any such injuries, and vests exclusive jurisdiction of the actions in the Southern District of New York. As for governing law, the Act provides that it shall be derived from the law, including choice of law principles, of the State in which the crash occurred unless such law is inconsistent with or preempted by Federal law. Finally, the Act provides that no air carriers liability shall exceed the limits of the carriers liability insurance coverage. The Acts provisions create some uncertainty as to the existence of federal question jurisdiction. On the one hand, as with the MMTJA, the trigger of jurisdiction under the ATSSSA does not appear to supply a substantial federal ingredient for any claims. But unlike the MMTJA, the ATSSSA does contain other hints at federal substance. The Act describes the action for damages as a federal cause of action. Further, the Act provides that the governing law is to be derived from state law; it does not simply require the application of state law. Notably, however, the statute gives little guidance to the federal courts as to how they should perform this derivation function, although it does suggest that the court should focus on the law of the three states where the crashes occurred. The fact that the Act also calls for some consideration of state choice of law rules suggests that it contemplates variation in the applicable law, depending on where the planes went down, and makes it difficult to argue that the federal courts should fashion a uniform body of federal common law to govern all claims. Its (seemingly unnecessary) affirmation that state law must yield when in conflict with federal law further underscores the Acts apparent assumption that state law generally controls. If the ATSSSA fails to specify a substantial federal ingredient that would clearly establish a basis for federal question jurisdiction, another jurisdictional possibility may exist. In the only clear articulation of a substantive rule, the Act imposes a cap on liability equal to the limits of the carriers insurance coverage. This cap may not provide an original federal ingredient for any individual suit brought under the Act, but it would provide a basis for the carriers to institute an interpleader action, joining all the potential claimants in a single proceeding. One might regard the cap as an original federal ingredient in such an interpleader proceeding. Alternatively, one might treat the interpleader action as coming within the district courts ancillary jurisdiction, so long as any adverse claimants satisfied the requirements of complete diversity. The grant of exclusive
175. Pub. L. No. 107-42, 115 Stat. 230 (2001).
1468
[Vol. 95:1423
jurisdiction in the Southern District of New York could be viewed as a venue provision, unambiguously fixing the location of the litigation and obligating the court to ensure that the liability cap is enforced. D. Evaluating the Consequences of Article III Limits
If inclined to impose Article III limits on multi-party litigation, the Court will doubtless consider the ability of Congress to use other tools to address perceived problems with state law class actions. The congressional toolkit would include at least two important alternative approaches. First, Congress could limit state court authority to entertain multi-state class actions on issues of state law.176 Congress could establish legislative limits on the state courts ability to exercise personal jurisdiction over the claims of individuals who have no contacts with the state. While rules of personal jurisdiction have largely developed without congressional guidance or involvement, few doubt that Congress has the power to regulate the circumstances in which a state court reaches beyond its borders to compel a party to join state court litigation. The crossing of state borders would implicate congressional power under the commerce clause, and the aggregate effect of multi-state and nationwide class actions would surely affect interstate commerce.177 Second, Congress has ample power to facilitate the adjudication of claims in federal court on a nationwide basis. If it were to choose this latter option, however, a Congress facing Article III constraints would be obligated to establish rules of federal substantive law under which such claims would arise. Some observers may view this requirement as unnecessarily disabling to a Congress that might simply wish to secure federal judicial control without facing the necessity of enacting a detailed code of consumer protection law. Current divisions over such issues as punitive and non-economic damages could make it difficult to reach legislative closure. One of the reasons Congress may prefer to shift disputes into federal court on the basis of state law is to avoid the legislative breakdown that the consideration of federal substantive rules
176. By adopting such a personal jurisdiction approach, Congress could address any perceived pathologies of state court class action litigation to which the expansive jurisdictional ruling of Shutts gave rise. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797 (1985). The Shutts decision upheld the power of the Kansas state court to entertain claims of class members from outside the state who lacked any affiliating connection to the state; the Court found that the failure of the class members to opt out of the litigation provided a sufficient basis for the state court to hear their claims. Id. at 811-12. As for the power of the state court to apply its own law to the claims of the non-resident class members, particularly those with no affiliating connection to the state, the Court applied established doctrine in holding that the state court may apply its law only to disputes with which the state has a significant contact or aggregation of contacts, creating state interests, such that the choice of law is neither arbitrary nor fundamentally unfair. Id. at 821-22 (citing Allstate Ins. Co. v. Hague, 449 302, 312-13 (1981)). 177. See id.
2007]
1469
seems to entail.178 Indeed, students of protective jurisdiction have cited the inability of Congress to legislate as one reason why it may prefer to shift matters to federal court for the application of state law.179 Such familiar federal tort programs as the Federal Tort Claims Act (FTCA) and the Foreign Sovereign Immunity Act rely on the incorporation of state rules of substance to define the extent of liability.180 But the Court may conclude that incorporation of state law as the standard of federal liability does not suffice, standing alone, to establish the federal substance needed to support the exercise of federal question jurisdiction. Note that in both the FTCA and FSIA, Congress incorporated state law as the measure of liability after making the federal government and foreign sovereigns legally responsible for their torts (and in the case of the FSIA, for their breaches of contract). The statutes do not simply incorporate existing state law standards of conduct; they subject government entities that had previously enjoyed sovereign immunity from suit to the same rules that govern the primary conduct of other parties. This change from immunity to accountability, with state law providing the measure of liability, represents an important substantive shift in legal relations. CAFA does not include any provisions that would effect a similar change in the nature of the underlying legal obligations. The Court might resist an attempt by Congress to address a lack of substance for federal question purposes simply by declaring that federal law incorporates state rules for decision. Mere incorporation of state law has a bootstrapping quality that makes it an unattractive vehicle for the exercise of federal question jurisdiction.181 Although the Court might defensibly conclude that Congress cannot respond to a decision invalidating CAFAs jurisdictional grant through the incorporation of state law, the Court may not demand much by way of
178. See American Law Institute, Complex Litigation Project 375 (1993) (acknowledging that national standards would simplify the resolution of complex litigation but doubting that the political process will yield the necessary consensus). 179. See Goldberg, supra note 1, at 576-77. 180. See 28 U.S.C. 1346(b) (establishing liability of the United States for the torts of its agents and employees and basing such liability on the law of the place where the act or omission occurred); 28 U.S.C. 1606 (making a foreign state that has engaged in commercial activity liable in the same manner and to the same extent as a private individual under like circumstances. 181. Professor Black reached a contrary conclusion, testifying that Congress could overcome any Article III problems with the consumer class action legislation of 1969 simply by incorporating state law standards of consumer protection into federal law. See Consumer Hearings, supra note 111, at 21, 25-27. Professor Black based the conclusion in part on a broad view of the scope of protective jurisdiction and in part on decisions that had upheld the power of Congress to incorporate state law as the standard of federal criminal liability in federal enclaves. See id. at 26 (citing US v. Sharpnack, 355 U.S. 286 (1958)). But the analogy does not persuade. State law does not apply of its own force within federal enclaves; incorporation of state law thus effects a substantive change in legal relations that differs from the federal incorporation of state consumer protection law to govern matters to which it already applies. Cf. Goldberg, supra note 1, at 555-56.
1470
[Vol. 95:1423
federal substance to justify the exercise of federal jurisdiction under Article III. Consider the Y2K Act of 1999, legislation that Congress adopted to regulate class actions seeking compensation for injuries resulting from the failure of computer programs to anticipate the arrival of year 2000.182 Although the Y2K Act includes its share of procedural provisions (including rules that require notice, particularized pleading, and mediation of disputes before litigation),183 it also contains some federal substance. The Act regulates the imposition of damages for economic losses, limits the award of punitive damages, and makes a variety of other changes in the substantive law.184 Any class action seeking damages under the Y2K Act would encounter these restrictions on available remedies.185 As in Verlinden, then, claims created by state law could nonetheless be said to arise under federal law; the claims would inevitably present remedial issues governed by federal law.186 Conclusion It may be too late to establish a workable limit on Congressional use of minimal diversity to expand the jurisdiction of federal courts. For the past generation or two, lawyers, academics and Supreme Court justices have understood that the decision in State Farm v. Tashire provides authority for grants of jurisdiction on the basis of minimal diversity between adverse claimants. The expansive language of Tashire, coupled with the Courts failure to identify the permissibly ancillary basis of the jurisdiction at issue there, have led us all to assume that Congress can freely transfer multi-party litigation to federal court so long as any two
182. Pub. L. No. 106-37, 106 Stat. 185 (1999) (codified at 15 U.S.C. 6601-6617). 183. See 15 U.S.C. 6606 (providing for pre-litigation notice and use of alternative dispute resolution techniques); Id. at 6607 (specifying rules of particularity that govern the pleading of claims under the statute). 184. See 15 U.S.C. 6604 (requiring proof in support of a punitive damages claim by clear and convincing evidence and limiting punitive damages to the lesser of three times compensatory damages or $250,000); Id. at 6605 (imposing a rule of proportionate liability for cases not involving contract claims and specifying rules to govern its determination); Id. at 6608 (imposing a duty on victims to mitigate damages); Id. at 6611 (barring recovery of damages for economic loss in tort actions, except in the case of intentional torts and those that result in tangible injury to property). 185. See 15 U.S.C. 6614 (providing for the assertion of original jurisdiction over any Y2K action brought as a class action). The structure of the Y2K Act assumes that state law will continue to provide the vehicle for suits brought to recover tort and contract damages for Y2K failures. Thus, 6603(b) provides that the Act creates no new cause of action and 6615 expressly preserves state law to the extent that it provides stricter limits on damages and liabilities that the Act itself. Yet despite the continuing relevance of state law in creating the cause of action, federal law will inevitably determine the viability of such claims and the extent of available damages. Such issues may or may not appear on the face of well-pleaded complaints under the Y2K Act, but inhere in the resolution of such claims. 186. The Court in Verlinden made clear that issues of federal substantive law, even though nominally arising as defenses to liability created by state law, would nonetheless make the action one arising under federal law within the meaning of Article III. See Verlinden B.V. v. Central Bank of Nigeria, 461 U.S. 480 (1983).
2007]
1471
opposing parties meet the diversity standard. CAFA and the MMTJA illustrate the breadth of the power that Tashire has been said to bestow on Congress. Yet the Court has struggled to preserve the apparent meaning of Article III as imposing limits on the scope of federal judicial power. On the federal question side of Article III, the Courts limits appear to have a modest bite in ruling out theories of protective jurisdiction that would permit Congress to transfer complex litigation to federal court for resolution in accordance with state law. Expansive reliance on minimal diversity responds in part to these limitations; the drafters of CAFA and the MMTJA chose a minimal diversity approach after doubts had arisen as to the constitutionality of earlier legislation that would have relied on protective jurisdiction to accomplish the same goals. With the growth of complex litigation, which inevitably involves at least some diversity of citizenship among some interested parties, minimal diversity poses a distinct threat to the Courts jurisprudence of limits. The Court might respond in two ways. First, it might simply agree that minimal diversity alone can ground the jurisdiction over multi-party litigation; such an approach should also imply some rethinking of the restrictions on the doctrine of protective jurisdiction. Alternatively, it might view Article III as the basis for a jurisprudence of limits on the diversity side of the docket. In extending the judicial power only to controversies between citizens of different states, Article III provides a straightforward but perhaps largely forgotten foundation for a modified version of the complete diversity rule. The Court might plausibly require every separate controversy to arise between opposing parties who satisfy the Article III standard as citizens of different states. Once such a diverse party controversy anchors the jurisdiction, Congress could certainly expand the scope of the jurisdiction to include other parties who claim an interest in the property or subject matter in dispute, claims that the Court has long allowed the lower courts to hear in the exercise of ancillary jurisdiction. Congress could thus provide for jurisdiction over disputes like Tashire; ancillary jurisdiction extends to the interpleader claims of additional parties who claim an interest in a property or fund brought before the federal court on the basis of a dispute between diverse parties. Such a separate controversy approach seems far from ideal. It exalts a somewhat formal notion of what constitute separate claims, embracing ancillary jurisdiction to expand a controversy between diverse citizens but declining the exercise of pendent-party jurisdiction over separate and nondiverse controversies. Further, it could threaten to invalidate the jurisdictional provisions of CAFA (and the MMTJA): the separate claims of non-diverse citizens joined in a Rule 23(b)(3) class action on the basis of common questions of law or fact (like claims joined on the basis that they
1472
[Vol. 95:1423
grow out of a mass disaster that triggers the MMTJA) do not appear to be ancillary to the district courts jurisdiction over the controversies between diverse parties. Yet, as the examples of the Airline Safety Act and Y2K Act reveal, Congress can provide for the federal judicial resolution of multi-party claims by providing rules of federal substance to govern liability and the extent of damages. The Court may conclude that limiting jurisdiction to disputes that implement such substantive federal law may best serve the long-term health of our system of judicial federalism. Perhaps that is the lesson of the Courts reluctance to embrace protective jurisdiction.

Protective Jurisdiction

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Protective Jurisdiction

Uploaded by

Copyright:

Available Formats

Protective Jurisdiction, Aggregate Litigation, and the Limits of Article III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

LIMITS OF ARTICLE III

CALIFORNIA LAW REVIEW

You might also like